1803y_En_Q22_Əyani_Yekun imtahan_ISE testinin sualları Fənn : 1803y Ekonometrika
1.
Which one of the following statements is true about quadratic associations?
√
It is not linear
√
It has threshold value
√
It captures decreasing or increasing marginal effects of a variable
√
It is U-shaped or reversed U-shaped
√
Functions that contain squares of one or more explanatory variables
√
they capture diminishing or increasing effects on the dependent variable
2.
Analyzing the behavior of unemployment rates across U.S. states in March of
2006 is an example of using
•
panel data
•
experimental data
√
cross-sectional data
•
time series data
3.
If you want to investigate income-consumption association among 17-65 age
group in Azerbaijan, collected data structure most likely will be:
•
Time-series data
•
Panel data
•
Pool data
•
Experimental data
√
Cross-sectional data
4.
Which of the following is true as stages of empirical analyses (step-by-step)?
•
Specification economic / econometric model, estimating parameters of
econometric model, statement of hypothesis, data collection.
√
Statement of hypothesis, specification economic / econometric model, data
collection, estimating parameters of econometric model.
•
Statement of hypothesis, specification economic / econometric model,
estimating parameters of econometric model, data collection
•
Specification economic / econometric model, data collection, statement of
hypothesis, estimating parameters of econometric model
5.
The parameter of an econometric model
✓
describe the strength of the relationship between the variable under study and
the factors affecting it
✓
constants of econometric models
6.
The constants of econometric models are referred to as
√
parameters
•
statistics
•
error terms
•
hypotheses
7.
The value of R2 (R-squared) is always
•
lies below 0
•
lies above 1
√
lies between 0 and 1
•
lies between 1 and 1.5
.
.
.
8.
Consider the following regression equation: y = β1 + β2x1 + β3x2 + u. What does
β2 imply?
•
independent variable
•
dependent variable
•
intercept parameter
√
slope parameter
√
β2 measures the ceteris paribus effect of x1 on y
9.
Which one of the following is true about econometrics?
√
Econometrics aims to analyze the relationship between X and Y in the population
•
All of them are true
•
Econometrics refer to theories and does not rely on empirical evidence
•
Econometrics is mix of mathematics and statistics
•
Econometrics aims to analyze the relationship between X and Y in the sample
10.
Which of the following is a difference between panel and pooled cross-sectional
data?
√
A panel data set consists of data on the same cross-sectional units over a given
period of time while a pooled data set consists of data on different crosssectional units over a given period of time
•
A panel data set consists of data on different cross-sectional units over a given
period of time while a pooled data set consists of data on the same crosssectional units over a given period of time.
•
A panel data consists of data on a single variable measured at a given point in
time while a pooled data set consists of data on the same cross-sectional units
over a given period of time.
•
A panel data set consists of data on a single variable measured at a given point in
time while a pooled data set consists of data on more than one variable at a
given point in time.
11.
Which of the following is true as stages of empirical analyses (step-by-step)?
•
Specification economic / econometric model, estimating parameters of
econometric model, hypothesis testing, interpretation of parameters, statement
of hypothesis, data collection.
√
Statement of hypothesis, specification economic / econometric model, data
collection, estimating parameters of econometric model, hypothesis testing,
interpretation of parameters
•
Statement of hypothesis, specification economic / econometric model, data
collection, hypothesis testing, estimating parameters of econometric model,
interpretation of parameters
•
Estimating parameters of econometric model, statement of hypothesis,
specification economic / econometric model, hypothesis testing, interpretation
of parameters data collection
•
Data collection, specification economic / econometric model, statement of
hypothesis, estimating parameters of econometric model, hypothesis testing,
interpretation of parameters
12.
One of the following statements best explain the difference between panel and
pooled cross-sectional data.
√
The key feature of panel data that distinguishes them from a pooled cross section
is that the same cross-sectional units (individuals, firms, or counties in the
preceding examples) are followed over a given time period.
√
A panel data set consists of data on the same cross-sectional units over a given
period of time while a pooled data set consists of data on different crosssectional units over a given period of time
13.
An empirical analyses includes,
√
An empirical analysis uses data to test a theory or to estimate a relationship
√
Empirical Analysis: A study that uses data in a formal econometric analysis to test
a theory, estimate a relationship, or determine the effectiveness of a policy.
14.
Which of the following refers to panel data?
•
Data on the price of a company’s share during a year.
•
Data on the unemployment rate in a country over a 5-year period
•
Data on the income of 5 members of a family on a particular year.
√
Data on the birth rate, death rate and population growth rate in developing
countries over a 10-year period.
15.
Which of the following is an example of time series data?
•
Data on the consumption of wheat by 200 households during a year.
•
Data on the number of vacancies in various departments of an organization on a
particular month.
√
Data on the gross domestic product of a country over a period of 10 years.
•
Data on the unemployment rates in different parts of a country during a year.
16.
A data set that consists of observations on a variable or several variables over
time is called a
√
Time series data set
•
experimental data set
•
panel data set
•
cross-sectional data set
17.
Data on the income of law graduates collected at a given point of time is
•
panel data
√
cross-sectional data
•
time series data
•
experimental data
.
18.
A data set that consists of a sample of individuals, households, firms, cities,
states, countries, or a variety of other units, taken at a given point in time, is
called
√
cross-sectional data set
•
pooled cross-sectional data set
•
time series data set
•
panel data set
19.
Which of the following is the first step in empirical economic analysis?
√
Statement of hypotheses
•
Testing of hypotheses
•
Collection of data
•
Specification of an econometric model
20.
The parameters of an econometric model
√
describe the strength of the relationship between the variable under study and
the factors affecting it
•
refer to the predictions that can be made using the model
•
include all unobserved factors affecting the variable being studied
•
refer to the explanatory variables included in the model
21.
The term ‘u’ in an econometric model is usually referred to as the
•
parameter
•
dependent variable
√
error term
•
hypothesis
.
.
22.
An empirical analysis relies on to test a theory.
√
data
•
customs and conventions
•
common sense
•
ethical considerations
23.
Which of the following is true of experimental data?
•
Experimental data is sometimes called retrospective data.
√
Experimental data are collected in laboratory environments in the natural
sciences.
•
Experimental data is sometimes called observational data.
•
Experimental data cannot be collected in a controlled environment.
24.
Econometrics is the branch of economics that
•
Applies mathematical methods to represent economic theories and solve
economic problems.
•
Deals with the performance, structure, behavior, and decision-making of an
economy as a whole
•
Studies the behavior of individual economic agents in making economic decisions
√
Develops and uses statistical methods for estimating economic relationships
25.
The t-statistic is calculated by dividing
•
the OLS estimator by its standard error.
•
the slope by the standard deviation of the explanatory variable.
√
the estimator minus its hypothesized value by the standard error of the
estimator.
•
the slope by 1.96.
26.
Analyzing the effect of minimum wage changes on teenage employment across
the 48 contiguous U.S. states from 1980 to 2004 is an example of using
•
cross-sectional data
•
having a treatment group vs a control group, since only teenagers receive
minimum wages
√
panel data
•
time series data
27.
Studying inflation in the United States from 1970 to 2006 is an example of using
•
cross-sectional data
•
randomized controlled experiments
√
time series data
•
panel data
28.
Consider X and Z variables, if the covariance value equals -0.65, this means:
•
There is positive association between X and Z
•
There is a strong association between X and Z
•
There is no relationship between X and Z
√
There is negative association between X and Z
29.
Consider the variables X and Y, standard deviation and variance of X show:
•
What is the direction of the relationship between X and Y
•
How much strong is the relationship between X and Y
•
How closely X values are distributed around mode of X
√
How closely X values are distributed around mean of X
•
How closely X values are distributed around median of X
30.
There are 25 marbles, each a different color in a bag. 3 of them is pink. What is
the probability of reaching in to the bag, without looking, and selecting a pink
marble?
•
25-3
√
0.12
•
7.333
•
0.136
•
8.333
31.
Consider X and Y variables, if the covariance value equals 0.45, this means:
√
There is positive association between X and Y
32.
The error term in a regression equation is said to exhibit homoskedasticty if
•
it has zero conditional mean
√
it has the same variance for all values of the explanatory variable.
•
it has the same value for all values of the explanatory variable
•
if the error term has a value of one given any value of the explanatory variable.
33.
In the equation Y = b0 + b1*X + u, Y is called as
√
Dependent variable
√
Explained variable
√
Response variable
√
Predicted variable
√
Regressand
34.
In the equation Y = b0 + b1*X + u, X is called as
√
Independent variable
√
Explanatory variable
√
Control variable
√
Predictor variable
√
Regressor
35.
R2 is the ratio of
√
explained variation compared to the total variation.
√
ESS/TSS,
√
explained sum of squared residuals by total sum of squared residuals,
√
SSE/SST
36.
Consider X and Y variables, if the covariance value equals 0.7, this means:
•
There is no relationship between X and Y
•
There is negative association between X and Y
•
There is a strong association between X and Y
√
There is positive association between X and Y
37.
Which one of the following statements is most likely true?
√
A variable whose value is determined by the outcome of a chance experiment is
called a random variable
•
A variable whose value is given by the outcome of a chance experiment is called a
binary variable
•
A variable whose value is given by the outcome of a chance experiment is called a
dependent variable
•
A variable whose value is given by the outcome of a chance experiment is called
an independent variable
38.
Standard deviation and variance show:
√
Deviations around the mean
•
Deviations from median of a variable
•
Direction of association between two variables
•
Strength of association between two variables
39.
Which one of the following statements is not true?
•
Sample is any subgroup of population
•
None of them
√
Smaller sample size is better than analyzing larger sample size for a research
•
Population covers all elements of target group
40.
If a number is added to a set that is far away from the mean, how does this affect
standard deviation?
•
Decrease significantly
•
Stay the same
√
Increase
•
Decrease slightly
41.
If the standard deviation of a data set is 4, what is the variance?
•
2
•
32
•
4
√
16
42.
There are 23 marbles, each a different color in a bag. One of them is pink. What is
the probability of reaching in to the bag, without looking, and selecting a pink
marble?
•
1/22
•
23/1
√
1/23
•
1
43.
The probability of an outcome
√
is the proportion of times that the outcome occurs
•
equals the sample mean divided by the sample standard deviation.
•
is the number of times that the outcome occurs
•
equals M × N, where M is the number of occurrences and N is the population size
44.
Ali examines income elasticity of consumption among youth (age between 17-35)
living of Baku. For this purpose, he conducts a survey among 1000 individuals
including 450 males and 550 females. According the given information, identify
the Population and Sample:
•
Population: all Baku citizens within age 17-35; Sample: 1000 Baku citizens
•
Population: 1000; Sample: 550
•
Population: all Baku citizens; Sample: 1000 Baku citizens
√
Population: all Baku citizens within age 17-35; Sample: 1000 Baku citizens with
age range of 17-35.
45.
An agency wants to know the opinions of Georgia residents on the construction
of Highway 316. The agency surveys 800 residents. 600 residents approve the
construction. 200 residents disapprove the construction. The agency concludes
that Georgia residents are pleased with the construction. According the given
information, identify the Population and Sample:
•
Population: 800 Residents; Sample: Georgia Residents
√
Population: Georgia Residents; Sample: 800 Residents
•
Population: Highway 316; Sample: 800 Residents
•
Population: 800 Residents; Sample: 200 Residents
46.
The sample covariance between the regressors and the Ordinary Least Square
(OLS) residuals is always .
√
zero
47.
A natural measure of the association between two random variables is
√
correlation coefficient
48.
The variance of the slope estimator as the error variance decreases.
√
decreases
.
49.
Consider the following regression model: y = β0 + β1x1 + u. Which of the
following is a property of Ordinary Least Square (OLS) estimates of this model
and their associated statistics?
•
The sum, and therefore the sample average of the OLS residuals, is positive.
•
The sum of the OLS residuals is negative.
•
The sample covariance between the regressors and the OLS residuals is positive.
√
The point ( x , y ) always lies on the OLS regression line.
50.
If x and y are positively correlated in the sample then the estimated slope is.
•
a. equal to zero
•
equal to one
•
less than zero
√
greater than zero
√
positive
51.
Azerbaijan State University of Economics (UNEC) randomly selected 150 UNEC
teachers to find out how much they are satisfied with university examination
policy. 90 teachers respond positively while 40 teachers expressed their
dissatisfaction and concerns. Remaining20 teachers are neither satisfied nor
dissatisfied with university examination policy. According to given information,
identify the sample size:
•
All teachers of UNEC
•
All teachers and students of UNEC
✓
150 teachers
•
90 teachers
52.
Azerbaijan State University of Economics (UNEC) randomly selected 150 UNEC
teachers to find out how much they are satisfied with university examination
policy. 90 teachers respond positively while 40 teachers expressed their
dissatisfaction and concerns. Remaining20 teachers are neither satisfied nor
dissatisfied with university examination policy. According to given information,
identify the population size:
•
All teachers and students of UNEC
✓
All teachers of UNEC
•
90 teachers
•
150 teachers
53.
In an econometric analysis, a population is:
✓
All individuals or any other items within the target group of a research
•
None of them
•
All individuals living in the target geographical area
•
Selected number of individuals within the target group of a research.
54.
Consider the variables X and Y, Cov(X,Y) show:
✓
Covariance measures the amount of linear dependence between two random
variables. A positive covariance indicates that two random variables move in the
same direction, while a negative covariance indicates they move in opposite
directions.
55.
The sum op probabilities of all possible outcomes to happen equals:
✓
1
56.
Consider X and Z variables, if the covariance value equals -0.70, this means:
•
There is positive association between X and Z
•
There is a strong association between X and Z
•
There is no relationship between X and Z
✓
There is negative association between X and Z
57.
The OLS residuals
•
can be calculated using the errors from the regression function
•
are unknown since we do not know the population regression function
✓
can be calculated by subtracting the fitted values from the actual values
•
should not be used in practice since they indicate that your regression does not
run through all your observations
58.
The regression R2 is a measure of (isarelenen iki cavab da duz olur)
✓
the goodness of fit of your regression line
✓
measures the strength of the relationship between your model and the
dependent variable on a convenient 0-100% scale.
•
whether or not X causes Y.
•
whether or not ESS > TSS. (TSS=ESS+RSS)
•
the square of the determinant of R.
59.
Which of the following statements is correct?
•
R2 = 1 - (ESS/TSS)
√
TSS = ESS + SSR
•
ESS = SSR + TSS
•
ESS > TSS
60.
The regression R2 (R-squared) is defined as follows: (hansi olsa secersiz)
✓
explained variation compared to the total variation.
✓
ESS/TSS,
✓
explained sum of squared residuals by total sum of squared residuals,
✓
SSE/SST
61.
Consider the given equation, Consumption = 80 + 0.8 * income. Note that
consumption and income are measured in AZN. How much is predicted value of
consumption if income equals 300 AZN?
•
300 AZN
•
240 AZN
✓
320 AZN
•
280 AZN
62.
Consider the given equation, Consumption = 120 + 0.8 * income. Note that
consumption and income are measured in AZN. Here what 0.8 means?
•
Ceteris paribus, in average, 1 %. increase in income increases consumption by 0.8
AZN.
•
Ceteris paribus, in average, 1 AZN increase in consumption increases
consumption by 0.8 AZN.
✓
Ceteris paribus, in average, 1 AZN increase in income increases consumption by
0.8 AZN.
•
Ceteris paribus, in average, 1 AZN increase in income increases consumption by
0.8 %.
63.
Consider the given equation, Consumption = 120 + 0.8 * income. Note that
consumption and income are measured in AZN. Here what 120 means?
•
Average value of consumption in the sample
✓
The value of consumption when income equals zero.
•
The value of income when consumption equals zero.
•
Average value of income in the sample
64.
Consider the given equation, Price = - 99001.1 + 2786 * Area. Note that “Price”
and “Area” denote price (measured in AZN) and largeness (measured in m2) of a
flat in Baku, respectively. If a house’s area equals 50 m2, its price is expected to
be:
•
99001.1 AZN
•
96215.1 AZN
•
139300 AZN
•
238301.1 AZN
✓
40298.9 AZN
65.
In a simple regression model, Y= b0 + b1*X + u, “u” is:
✓
Difference between actual value and fitted value of the dependent variable
•
Difference between actual value and fitted value of an explanatory variable.
•
Difference between explained and unexplained part in the model
•
Difference between actual value of dependent variable and fitted value of the
independent variable.
•
Difference between fitted value and estimated value of dependent variable
66.
Which one is the difference between sample regression function (SRF) and and
population regression function (PRF)?
•
PRF parameters can be estimated according to randomly selected sample data
and generalized to SRF
•
SRF includes only one independent variable while PRF may include many
independent variables
✓
SRF parameters can be estimated according to randomly selected sample data
and generalized to PRF
•
PRF is used to have an idea about the relationship in SRF
•
None of them
67.
Which of the following is assumed for establishing the unbiasedness of Ordinary
Least Square (OLS) estimates?
•
The error term has an expected value of 1 given any value of the explanatory
variable.
•
The sample outcomes on the explanatory variable are all the same value.
✓
The regression equation is linear in terms of parameters
•
The error term has the same variance given any value of the explanatory variable.
68.
If the residual sum of squares (SSR) in a regression analysis is 66 and the total
sum of squares (SST) is equal to 90, what is the value of the coefficient of
determination (goodness of fit)?
✓
0,27
•
1,2
•
0,73
•
0,55
69.
If the total sum of squares (SST) in a regression equation is 81, and the residual
sum of squares (SSR) is 25, what is the explained sum of squares (SSE)?
✓
56
•
18
•
64
•
32
70.
In the equation Y = b0 + b1*X + u, which of the following is a property of Ordinary
Least Square (OLS) estimates of this model and their associated statistics?
•
The sum, and therefore the sample average of the OLS residuals, is positive.
•
The sample covariance between the regressors and the OLS residuals is positive.
✓
The sum, and therefore the sample average of the OLS residuals, is zero.
•
The sum of the OLS residuals is negative
71.
In the equation Consumption = b0 + b1 * Income + u, what is the residual for the
5th observation is actual value of consumption is 500AZN and fitted value of
consumption is 475 AZN?
✓
25 AZN
•
975 AZN
•
300 AZN
•
50 AZN
72.
Which statement best represent the difference between population regression
function (PRF) and sample regression function (SRF)?
•
PRF less accurately represent the association between an independent and
dependent variable compared to SRF.
•
None of them.
✓
PRF represent the true model in the population while SRF is obtained from
selected sample.
•
We reach to SRF based on parameters of PRF
73.
The residual is:
•
Difference between actual value and fitted value of an explanatory variable.
•
Difference between actual value of dependent variable and fitted value of
independent variable.
•
Difference between fitted value and estimated value of dependent variable
✓
Difference between actual value and fitted value of the dependent variable
74.
In the equation Y = b0 + b1*X + u, u is called as
•
independent variable
•
intercept parameter
✓
residual
•
slope parameter
75.
In the equation Y = b0 + b1*X + u, b1 is called as
✓
slope parameter
•
residual
•
independent variable
•
intercept parameter
76.
In the equation Y = b0 + b1*X + u, b0 is called as
•
slope parameter
•
dependent variable
•
independent variable
✓
intercept parameter
77.
If a change in variable x causes a change in variable y, variable x is called the ___.
•
dependent variable
•
explained variable
✓
explanatory variable
•
response variable
78.
A dependent variable is also known as a(n) _____.
•
independent variable
•
explanatory variable
•
control variable
✓
explained variable
79.
In the equation saving = b0 + b1 * Income + u, what is the residual for the 9th
observation is actual value of saving is 450 AZN and fitted value of saving is 420
AZN?
✓
30
80.
In the equation saving = b0 + b1 * Income + u, what is the residual for the 4th
observation is actual value of saving is 370 AZN and fitted value of saving is 330
AZN?
✓
40
81.
In the equation saving = b0 + b1 * Income + u, what is the residual for the 6th
observation if actual value of saving is 650 AZN and fitted value of saving is 490
AZN?
✓
160
82.
If the residual sum of squares (SSR) in a regression analysis is 42 and the total
sum of squares (SST) is equal to 73, what is the value of the coefficient of
determination?
✓
0.42, 31/73
83.
If the residual sum of squares (SSR) in a regression analysis is 60 and the total
sum of squares (SST) is equal to 95, what is the value of the coefficient of
determination?
✓
0.37, 35/95
84.
If the residual sum of squares (SSR) in a regression analysis is 54 and the total
sum of squares (SST) is equal to 87, what is the value of the coefficient of
determination?
✓
0.38, 33/87
85.
If the total sum of squares (SST) in a regression equation is 97, and the residual
sum of squares (SSR) is 46, what is the explained sum of squares (SSE)?
✓
51
86.
If the total sum of squares (SST) in a regression equation is 54, and the residual
sum of squares (SSR) is 17, what is the explained sum of squares (SSE)?
✓
37
87.
If the total sum of squares (SST) in a regression equation is 70, and the residual
sum of squares (SSR) is 39, what is the explained sum of squares (SSE)?
✓
31
88.
Consider the given equation, Price = - 99001.1 + 2786 * Area. Note that “Price”
and “Area” denote price (measured in AZN) and largeness (measured in m2) of a
flat in Baku, respectively. If a house’s area equals 70 m2, its price is expected to
be:
✓
96018.9
89.
Consider the given equation, Consumption = 90 + 0.75 * income. Note that
consumption and income are measured in AZN. Here 90 represent:
•
Average value of consumption in the sample
✓
The value of consumption when income equals zero.
•
The value of income when consumption equals zero.
•
Average value of income in the sample
90.
In a simple regression model, Y= b0 + b1*X + u,
✓
two-variable linear regression model or bivariate linear regression model
91.
Consider given model of the relationship between inflation and number of
unemployed people in Azerbaijan. Log(unemployed) = 1.9 +0.095 *inflation. Note
that inflation is measured in percent and number of unemployed people is
measured as thousands of people. Estimated model is:
•
Elasticity model
•
Inverse semi-elasticity (level-log)
✓
Semi-elasticity model
•
Level-level (lin-lin) model
92.
Assume that you have collected a sample of observations from over 100
households and their consumption and income patterns. Using these
observations, you estimate the following regression Ci = β0+β1Yi+ ui where C is
consumption and Y is disposable income. The estimate of β1 will tell you
•
ΔIncome/Δconsumption
✓
ΔConsumption/Δincome
•
The amount you need to consume to survive
•
Consumption/Income
•
None of them
93.
A type II error
✓
is the error you make when not rejecting the null hypothesis when it is false.
✓
failing to reject false null hypothesis
94.
Type I error is
✓
the error you make when rejecting the null hypothesis when it is true.
✓
rejecting true null hypothesis
95.
Consider given model of the relationship between inflation and number of
unemployed people in Azerbaijan. Log(unemployed) = 1.9 +0.095 *inflation. Note
that inflation is measured in percent and number of unemployed people is
measured as thousands of people. Here, 0.095 means?
•
In average, ceteris paribus, 1%. increase in inflation leads to 9.5 thousand
increase in amount of unemployed people.
•
In average, ceteris paribus, 1%. increase in inflation leads to 0.095%. increase in
amount of unemployed people.
•
In average, ceteris paribus, 1%. increase in inflation leads to 0.095 thousand
increase in amount of unemployed people.
✓
In average, ceteris paribus, 1%. increase in inflation leads to 9.5%. increase in
amount of unemployed people.
96.
Consider given model of the relationship between inflation and number of
unemployed people in Azerbaijan. Log(unemployed) = 5.3 +0.015*inflation. Note
that inflation is measured in percent and number of unemployed people is
measured as thousands of people. Here, 0.015 means?
✓
In average, ceteris paribus, 1%. increase in inflation leads to 1.5%. increase in
amount of unemployed people.
•
In average, ceteris paribus, 1%. increase in inflation leads to 0.015%. increase in
amount of unemployed people.
•
In average, ceteris paribus, 1%. increase in inflation leads to 0.015 thousand
increase in amount of unemployed people.
•
In average, ceteris paribus, 1%. increase in inflation leads to 1.5 thousand
increase in amount of unemployed people.
97.
Consider given model of the relationship between inflation and number of
unemployed people in Azerbaijan. Inflation = 4.3 +0.085*unemployed. Note that
inflation is measured in percent and number of unemployed people is measured
as thousands of people. Here, 0.085 means?
•
In average, ceteris paribus, 1%.increase in total number of unemployed
individuals leads 0.0517%. increase in inflation.
•
In average, ceteris paribus, 1%.increase in total number of unemployed
individuals leads 8.5%. increase in inflation.
•
In average, ceteris paribus, 1%.increase in total number of unemployed
individuals leads 8.77%. increase in inflation.
✓
In average, ceteris paribus, 1 thousand people increase in total number of
unemployed individuals leads 0.085%. increase in inflation.
98.
Consider given model of the relationship between inflation and number of
unemployed people in Azerbaijan. Inflation = 3.6 +5.17*log(unemployed). Note
that inflation is measured in percent and number of unemployed people is
measured as thousands of people. Here, 5.17 means?
✓
In average, ceteris paribus, 1% increase in total number of unemployed leads to
0.0517% increase in inflation
99.
Consider given two models estimated by using the same samples. First model is
log(income) = 4.1 + 0.125*work_hour. Note that income represents total job
earnings of an employee, and work hour represents total working hours of the
same employee. If goodness of fit for the model is 0.78, which statement is true?
•
Less than half of variation in working hours is explained by job earnings.
✓
More than half of variation in job earnings is explained by working hours.
•
More than half of variation in working hours is explained by job earnings.
•
Less than half of variation in job earnings is explained by working hours.
100. Consider a given model, log(GPA)=3.6 + 0.078*hours where GPA represent
average cumulative score of a student and Hours represent the amount of
average weekly studying hours. If goodness of fit for the model is 0.46, which
statement is true?
•
GPA explains 46%. of variation in studying hours
•
Studying hours explain 0.46%. of variation in GPA.
•
GPA explains 0.46%. of variation in studying hours
✓
Studying hours explain 46%. of variation in GPA.
101. Consider a given model, log(GPA)=3.6+0.078*hours where GPA represent
average cumulative score of a student and Hours represent the amount of
average weekly studying hours. According to the model, which of the following
statements is true?
√
1%. increase in studying hours is predicted to increase GPA by 7.8%., in average,
ceteris paribus.
•
There is negative association between studying hours and GPA of students.
•
1%. increase in studying hours is predicted to increase GPA by 0.078%., in
average, ceteris paribus.
•
1%. increase in GPA is predicted to increase studying hours by 0.078%., in
average, ceteris paribus.
102. Consider given two models estimated by using the same samples. First model is
log(income)=3.2+0.06*work_hour. The second model is
log(income)=4.1+0.36*log(work_hour). Goodness of Fit (R-squared) indicator is
0.42 for the first model, and 0.65 for the second model. Which model is better to
rely on, and why?
•
Given information does not allow to make any comparison between given models
•
Given information does not allow to make any comparison between given models
•
First model is better which is more precise, explain larger variation in dependent
variable.
•
Second model is better in which coefficient of Work_hour is greater than the
corresponding coefficient in the first model.
√
Second model is better which is more precise, explain larger variation in
dependent variable.
103. Consider the given equation, Log(GPA)=4.5+0.16*log(ent score). Note that GPA
represent cumulative grade point average score, and ent_score stand for
university entrance score of a student. According to the estimated model:
√
Ceteris paribus, in average, 1 percent increase in entrance score increases GPA
score by 0.16 percent.
•
Ceteris paribus, in average, 1 point increase in entrance score increases GPA
score by 0.0016 percent.
•
Ceteris paribus, in average, 1 point increase in entrance score increases GPA
score by 0.16 percent.
•
Ceteris paribus, in average, 1 percent increase in entrance score increases GPA
score by 0.0016 point
104. Consider the given equation, Log(GPA)=6.2+0.0056*ent score. Note that GPA
represent cumulative grade point average score, and ent_score stand for
university entrance score of a student. According to the estimated model:
•
Ceteris paribus, in average, 1 percent increase in entrance score increases GPA
score by 0.0056 point.
•
Ceteris paribus, in average, 1 percent increase in entrance score increases GPA
score by 0.0056 percent.
•
Ceteris paribus, in average, 1 point increase in entrance score increases GPA
score by 0.0056 percent.
√
Ceteris paribus, in average, 1 point increase in entrance score increases GPA
score by 0.56 percent.
105. In which model, semi-elasticity association has been estimated. Note that GPA
represent cumulative grade point average score, and ent_score stand for
university entrance score of a student.
✓
Log(GPA)= 6.2 + 0.056*ent_score
•
GPA = 35 + 0.2*ent_score
•
GPA = 41 + 20.5*log(ent_score)
•
Log(GPA) = 5.7 + 0.5*log(ent_sco
106. In which model, elasticity association has been estimated. Note that GPA
represent cumulative grade point average score, and ent_score stand for
university entrance score of a student.
•
GPA = 41 + 20.5*log(ent_score)
•
GPA = 35 + 0.2*ent_score
•
Log(GPA)= 6.2 + 0.0056*ent_score
✓
Log(GPA) = 5.7 + 0.5*log(ent_score)
107. Consider the given equation, log(GPA)=5.1+0.3*log(ent_score). Note that GPA
represent cumulative grade point average score, and ent_score stand for
university entrance score of a student. According to the estimated model:
•
There is negative association between university entrance score and GPA.
•
GPA has positive effect on entrance score of a student.
•
Ceteris paribus, in average, 1 point increase in entrance score increases GPA 0.3
point.
√
Ceteris paribus, in average, 1%. increase in entrance score increases GPA 0.3 %.
108. Consider the given equation, Consumption=80+0.8*income. Note that
consumption and income are measured in AZN. How much consumption is
predicted to change if income increases 200 AZN?
•
140 AZN
•
240 AZN
•
180 AZN
√
160 AZN
109. Consider the given equation, Price=-931284.8+247695.5*log(Area). Note that
“Price” and “Area” denote price(measured in AZN) and largeness
(measuredinm2) of a flat in Baku, respectively. According to the given equation,
which one of the following statements is most likely true:
•
Ceteris paribus, house price is expected to increase 1% when area increases
247695.5 unit.
√
In average, while holding other factors fixed, if area increases 1%, price of the
house is expected to increase 2476.955 AZN.
•
Ceteris paribus, house price will increase 247695.5 AZN when area increases 1 m2
•
Ceteris paribus, if area increases 1%, price of the house is expected to increase
2476.955 AZN.
•
In average, if area increases 1%, price of the house will increase 247695.5 AZN.
110. Consider the given equation, log(Price)=7.27+1.022*log(Area). Note that “Price”
and “Area” denote price (measured in AZN) and largeness(measured in m2) of a
flat in Baku, respectively. According to the given equation, which one of the
following statements is most likely true:
•
Ceteris paribus, house price is expected to increase 1% when area increases
1.022%
•
Ceteris paribus, in average, house price will increase 1 % when area increases
1.022 m2.
•
Ceteris paribus, in average, house price will increase 1.022% when area increases
1 m2.
√
In average, while holding other factors fixed, if area increases 1%, price of the
house is expected to increase 1.022%
•
In average, if area increases 1%, price of the house will increase 1.022%
111. Consider the given equation, log(Price)=11.03+0.008*Area. Note that “Price” and
“Area” denote price (measured in AZN) and largeness (measured in m2) of a flat
in Baku, respectively. Here, “0.008” means:
•
Ceteris paribus, in average, if area increases 1m2, price will increase 0.008%
•
Ceteris paribus, in average, if area increases 1%, price will increase 0.008%
•
Ceteris paribus, in average, if price increases 1%, area will increase 0.008 m2.
•
Ceteris paribus, in average, if price increases 1%, area will increase 0.8 %.
√
Ceteris paribus, in average, if area increases 1m2, price will increase 0.8%
112. If R2ur(R-squared (unrestricted))=0.8214, R2r(R-squared(restricted))=0.5968,
number of restrictions=5, and n–k–1=225, F statistic equals:
Formula bu 4 misal üçün: 𝐹h𝑒𝑠 = (𝑅1^2 − 𝑅2^2 )⁄𝑞 (1 − 𝑅1^2 )/(𝑛 − 𝑘 − 1)
✓
56.59
113. If R2ur(R-squared(unrestricted))=0.7273, R2r(R-squared(restricted))=0.5768,
number of restrictions=2,and n–k–1=150, F statistic equals:
✓
41.39
114. If R2ur(R-squared(unrestricted))=0.7184, R2r(R-squared(restricted))=0.5650,
number of restrictions=4, and n–k–1=210, F statistic equals:
✓
28.60
115. If R2ur(R-squared(unrestricted))=0.6873, R2r(R-squared(restricted))=0.5377,
number of restrictions=3, and n–k–1=229, F statistic equals:
✓
36.52
116. Which of the tests is used to test single variable
✓
t-test
117. The term “linear” in a multiple linear regression model means that the equation
is linear in
✓
parameters
118. If you exclude a relevant variable from a multiple linear regression model, you
will face the problem of.
•
multicollinearity
•
perfect collinearity
•
homoskedasticity
✓
omitted variable biasedness
✓
misspecification of the model
119. Consider the given equation, Price=-785232.8+135820.5*log(Area). Note that
“Price” and “Area” denote price(measured in AZN) and largeness(measured in
m2) of a flat in Baku, respectively. According to the given equation, which one of
the following statements is most likely true:
√
In average, while holding other factors fixed, if area increases 1%, price of the
house is expected to increase 1358.205 AZN.
√
Ceteris paribus, in average, if area increases 1%, price of the house is expected to
increase 1358.205 AZN.
120. Consider the given equation, log(Price)=6.25+1.23*log(Area). Note that “Price”
and “Area” denote price(measured in AZN) and largeness(measured in m2) of a
flat in Baku, respectively. According to the given equation, which one of the
following statements is most likely true:
√
In average, while holding other factors fixed, if area increases 1%, price of the
house is expected to increase 1.23%
√
Ceteris paribus, in average, if area increases 1%, price of the house is expected to
increase 1.23%
121. Consider the given equation, log(Price)=11.88-0.035*Distance. Note that “Price”
and “Distance” denote price(measured in AZN) and distance from the nearest
metro station(measured in km) of a flat in Baku, respectively. Here,“0.035”means
✓
In average, ceteris paribus, when distance increases 1 km, price is expected to
decrease 3.5%
122. Gauss-Markov assumptions do not include:
✓
Normality assumption
•
No perfect collinearity assumption
•
Random sampling assumption
•
Homoscedasticity assumption
123. If 100% of variation in an independent variable is explained by other independent
variables of the regression model:
•
Error variance will be biased
•
Model parameters will be unbiased
•
Model parameters will be biased
√
Estimation of the parameters and error variance is impossible.
124. Consider the given model, GDP=b0+b1*Total_trade+b2*Exports+b3*Imports+u.
Because, total trade = imports + exports, the model most likely violates:
•
Homoscedasticity assumption.
•
Zero conditional mean assumption
•
Normality assumption
√
No perfect collinearity assumption
125. If an independent variable has no variation,i.e.,all values are the same.
•
Estimated model parameters will be biased.
•
Estimated model will have serial-correlation problem
√
The model cannot be estimated.
•
Estimated model will have heteroscedasticity problem
126. Linear in parameters assumption requires that:
•
The relationship between independent variables and dependent variables must
be linear.
•
None of them.
√
Parameters of the model must be linear, i.e., numerical values.
•
Model should have linear relationship specification
127. The normality assumption implies that
•
The population error u is dependent on the explanatory variables and is normally
distributed with mean equal to one and variance σ2.
•
The population error u is dependent on the explanatory variables and is normally
distributed with mean zero and variance σ.
√
The population error u is independent of the explanatory variables and is
normally distributed with mean zero and variance σ2.
•
The population error u is independent of the explanatory variables and is
normally distributed with mean equal to one and variance σ.
128. Which one of the following statements is false?
•
The term “linear” in a multiple linear regression model means that the equation
is linear in parameters.
√
All factors in the unobserved error term be correlated with the explanatory
variables.
•
An explanatory variable is said to be exogenous if it is not correlated with the
error term.
•
The coefficient of determination (R2) increases when an independent variable is
added to a multiple regression model
129. Which one of the following statements is true?
•
The coefficient of determination (R2) decreases when an independent variable is
added to a multiple regression model.
•
An explanatory variable is said to be exogenous if it is correlated with the error
term.
•
All factors in the unobserved error term be correlated with the explanatory
variables.
√
The term “linear” in a multiple linear regression model means that the equation
is linear in parameters.
130. Sample error variance is supposed to be unbiased estimator of population error
variance, if?
✓
All of them
•
Sampling is done randomly
•
Expected value of error term is uncorrelated with explanatory variables
•
Residuals have equal variance at different values of the independent variable
131. For unbiasedness of a single regression model parameter:
•
All of them
•
Residuals must have normal distribution
•
Homoscedasticity assumption must be hold
✓
There must be variation in the explanatory variable value
132. One of the following assumptions is not required for unbiasedness of regression
parameters?\
•
Zero conditional mean assumption
•
Random sampling
•
Linearity in parameters
✓
Homoscedasticity
133. The Gauss-Markov theorem will not hold if
•
the error term has the same variance given any values of the explanatory
variables
•
the regression model relies on the method of random sampling for collection of
data
✓
the independent variables have exact linear relationships among them
•
the error term has an expected value of zero given any values of the independent
variables
134. High(but not perfect) correlation between two or more independent variables is
called
•
Perfect collinearity
•
heteroskedasticty
•
homoskedasticty
✓
multicollinearity
135. In the regression of y on x, the error term exhibits heteroscedasticity if
•
Var(y|x) is a function of x
•
y is a function of x
✓
it has not a constant variance
•
x is a function of y
136. The error term in a regression equation is said to exhibit homoscedasticty if
✓
it has the same variance for all values of the explanatory variable.
•
if the error term has a value of one given any value of the explanatory variable.
•
it has zero conditional mean
•
it has the same value for all values of the explanatory variable
137. One of the following assumptions is required for unbiasedness of error variance?
•
Random sampling
•
Homoscedasticity
•
Sample variation in explanatory variables
✓
All of them
•
Linear in parameters
138. One of the following assumptions is required for unbiasedness of regression
parameters?
✓
All of them
•
Sample variation in explanatory variables
•
No perfect collinearity
•
Linear in parameters
•
Random Sampling
139. Ordinary Least Squares method estimates parameters which:
•
Minimizes the total variation in fitted values
•
Maximizes the goodness of fit indicator
•
Supports the expectations according to economic theory
✓
Minimizes the sum of squared residuals
•
Minimizes sum of residuals
140. If the model satisfies the first four Gauss-Markov assumptions, then v has
•
a zero mean and is correlated with only x1
•
a zero mean and is correlated with x1and x2
•
a zero mean and is correlated with only x2
✓
a zero mean and is uncorrelated with x1 and x2
141. If Cov(z,x) ≠ 0, then z and x are _____
✓
correlated
142. One of the following statements is not true
143. Homoscedasticity assumption requires that
✓
the variance of u cannot depend on any element of X, and the variance must be
constant across observations
The value of the explanatory variable must contain no information about the
variability of the unobserved factors.
144. One of the following assumptions is not required for unbiasedness of regression
parameters?
•
Zero conditional mean assumption
•
Random sampling
•
Linearity in parameters
√
Homoscedasticity
145. Suppose the variable x2 has been omitted from the following regression
equation, y=β0+β1 x1+β2 x2+u. Also, note that α1 is the estimator obtained
when x2 is omitted from the equation. If α1> β1, α1 is said to
•
Be biased toward zero.
√
Have a positive bias
•
Be unbiased
•
Have a negative bias
146. Suppose the variable x2 has been omitted from the following regression
equation, y=β0+β1 x1+β2 x2+u. Also, note that α1 is the estimator obtained
when x2 is omitted from the equation. α1 will be equal to β1 if
•
β2 >0 and x 1 and x 2 are positively correlated
•
β2 <0 and x 1 and x 2 are negatively correlated
√
β2 = 0 and x 1 and x 2 are negatively correlated
•
β2 <0 and x 1 and x 2 are positively correlated
147. Suppose the variable x2 has been omitted from the following regression
equation, y=β0+β1 x1+β2 x2+u. Also, note that α1 is the estimator obtained
when x2 is omitted from the equation. The bias in α1 is negative
(underestimated) if
•
β2 = 0 and x 1 and x 2 are negatively correlated
•
β2 <0 and x 1 and x 2 are negatively correlated
•
β2 >0 and x 1 and x 2 are positively correlated
√
β2 <0 and x 1 and x 2 are positively correlated
148. Suppose the variable x2 has been omitted from the following regression
equation, y=β0+β1 x1+β2 x2+u. Also, note that α1 is the estimator obtained
when x2 is omitted from the equation. The bias in α1 is positive (overestimated)
if
•
β2 = 0 and x 1 and x 2 are negatively correlated
•
β2 <0 and x 1 and x 2 are positively correlated
•
β2 >0 and x 1 and x 2 are negatively correlated
√
β2 >0 and x 1 and x 2 are positively correlated
149. Exclusion of a relevant variable from a multiple linear regression model leads to
the problem of
•
Multicollinearity
•
Irrelevant variable biasedness
√
Omitted variable biasedness
•
Perfect collinearity
150. If an independent variable in a multiple linear regression model is an exact linear
combination of other independent variables, the model suffers from the problem
of
•
omitted variable bias
•
heteroskedasticty
•
homoskedasticity
√
perfect collinearity
151. The value of R-squared (goodness of fit) always
√
is between 0 and 1
•
is between 1 and 1.5
•
is greater than 1
•
is less than 0
152. Which of the following is true of R-squared (goodness of fit)?
•
R-squared usually decreases with an increase in the number of independent
variables in a regression.
√
R-squared shows what percentage of the total variation in the dependent
variable, Y, is explained by the explanatory variables.
•
A low R-squared indicates that the Ordinary Least Squares line fits the data well.
•
R-squared is also called the standard error of regression.
153. If the explained sum of squares is 35 and the total sum of squares is 49, what is
the residual sum of squares?
•
10
•
18
√
14
•
12
154. In the equation, Y=b0 + b1*X1 + b2 * X2 + u, what does b1 imply?
•
b1 measures the ceteris paribus effect of X1 on U
•
b1 measures the ceteris paribus effect of Y on X1
•
b1 measures the ceteris paribus effect of X1 on X2
√
b1 measures the ceteris paribus effect of X1 on Y
155. In the equation, Y=b0 + b1*X1 + b2 * X2 + U, b2 is a(n)
√
slope parameter
•
intercept parameter
•
independent variable
•
dependent variable
156. Which of the following correctly identifies an advantage of using adjusted R2 over
R2?
√
The penalty of adding new independent variables is better understood through
adjusted R2 than R2.
•
The adjusted R2 can be calculated for models having logarithmic functions while
R2 cannot be calculated for such models.
•
Adjusted R2 corrects the bias in R2.
•
Adjusted R2 is easier to calculate than R2
157. In a multiple regression model, an irrelevant variable is considered to be:
√
Including or not including an irrelevant variable affects variance of parameters
while slope parameters does not change
•
Including or not including an irrelevant variable affects goodness of fit indicator
while slope parameters and variance of parameters does not change
•
Including or not including an irrelevant variable makes model parameters biased
•
Including or not including an irrelevant variable does not affect regression results
at all
•
Including or not including an irrelevant variable affects slope parameters of the
regression model while goodness of fit indicator does not change
158. Consider the given equation, Income=102.5+540.1 *log(work_hour)+3.26 *Exper.
Note that income denote monthly salary, and work_hour denote average
working hours per week while Exper represent total job experience (measured in
years) of the individual. In the model, 540.1 means?
•
Work hours explain 54.01% of variation in income.
•
In average, ceteris paribus, 1 hour increase in weekly working hours is expected
to decrease income by 540.1 AZN
•
In average, ceteris paribus, 1 hour increase in weekly working hours is expected
to increase income by 540.1 AZN.
•
In average, ceteris paribus, 1 % increase in weekly working hours is expected to
increase income by 5.401 %.
√
In average, ceteris paribus, 1 % increase in weekly working hours is expected to
increase income by 5.401. AZN.
159. Consider the given equation, Y = b0 + b1*X1 + b2 * X2 + u. Here, what does “b1”
imply?
✓
b1 measures the ceteris paribus effect of X1 on Y
•
b1 measures the ceteris paribus effect of X1 on U
•
b1 measures the ceteris paribus effect of X1 on X2
•
b1 measures the ceteris paribus effect of Y on X1
160. In the following equation, Import refers to Import Demand, NI refers National
Income, and CPI refers the Consumer Price Index. log(Import) = 5.93 +
0.73*log(NI) + 0.244*CPI. Which of the following statements is then true?
✓
In average, if National Income increases by 1 percent, Import increases by
approximately 0.73%., the amount of CPI remaining constant.
✓
In average, if CPI increases by 1 unit, Import increases by approximately 24.4%,
the amount of National Income remaining constant.
161. In the following equation, Import refers to Import Demand, NI refers National
Income, and CPI refers the Consumer Price Index. log(Import) = 7.15 + 0.5*log(NI)
+ 0.46*CPI. Which of the following statements is then true?
✓
In average, if National Income increases by 1 percent, Import increases by
approximately 0.5%., the amount of CPI remaining constant.
✓
In average, if CPI increases by 1 unit, Import increases by approximately 46%, the
amount of National Income remaining constant.
162. In the following equation, Import refers to Import Demand, NI refers National
Income, and CPI refers the Consumer Price Index. log(Import) = 1.17 +
0.643*log(NI) + 0.317*CPI. Which of the following statements is then true?
✓
In average, if National Income increases by 1 percent, Import increases by
approximately 0.643%., the amount of CPI remaining constant.
✓
In average, if CPI increases by 1 unit, Import increases by approximately 31.7%,
the amount of National Income remaining constant
163. In the following equation, Import refers to Import Demand, NI refers the National
Income and CPI refers Consumer Price Index. log(Import) = 1.95 + 0.27*log(NI) +
0.355*CPI. Which of the following statements is then true?
✓
In average, if National Income increases by 1 percent, Import increases by
approximately 0.27%., the amount of CPI remaining constant.
✓
In average, if CPI increases by 1 unit, Import increases by approximately 35.5%,
the amount of National Income remaining constant
164. In the following equation, Import refers to Import Demand, NI refers the National
Income and CPI refers Consumer Price Index. log(Import) = 2.49 + 0.627*log(NI) +
0.496*CPI. Which of the following statements is then true?
✓
In average, if National Income increases by 1 percent, Import increases by
approximately 0.627%., the amount of CPI remaining constant.
✓
In average, if CPI increases by 1 unit, Import increases by approximately 49.6%,
the amount of National Income remaining constant.
165. In the following equation, Import refers to Import Demand, NI refers the National
Income and CPI refers Consumer Price Index. log(Import) = 5.33 + 0.437*log(NI) +
0.355*CPI. Which of the following statements is then true?
✓
In average, if National Income increases by 1 percent, Import increases by
approximately 0.437%., the amount of CPI remaining constant.
✓
In average, if CPI increases by 1 unit, Import increases by approximately 35.5%,
the amount of National Income remaining constant.
✓
National income and Import Demand … model is elasticity (log-log)
✓
CPI and Import is semi elasticity (log-lin)
166. Consider that true model is y = b0 + b1*x1 + b2*x2 +u. If “x2” is not included to
the estimated model, then the estimated model most likely has:
✓
omitted variable biasedness,
✓
misspecification of the model
167. Consider the given equation, Income=102.5+540.1 *log(work_hour)+3.26 *Exper.
Note that income denote monthly salary, and work_hour denote average
working hours per week while Exper represent total job experience (measured in
years) of the individual. How much income is predicted to change when weekly
working hours increase 5% while job experience decreases 2 years?
✓
20.485
168. Which of the following statements is true about R-squared (goodness of fit)?
•
R-squared usually decreases with an increase in the number of independent
variables in a regression.
√
R-squared shows what percentage of the total variation in the dependent
variable, Y, is explained by the explanatory variables.
•
A low R-squared indicates that the Ordinary Least Squares line fits the data well.
•
R-squared is also called the standard error of regression
169. In the following equation, GDP refers to gross domestic product, and FDI refers
the foreign direct investment. log(gdp) = 2.65 + 0.527*log(bankcredit) +
0.222*FDI. Which of the following statements is then true?
•
In average, if FDI increases by 1 unit (1 million USD), GDP increases by
approximately 0.222%., the amount of bank credit remaining constant.
•
In average, if FDI increases by 1 unit (1 million USD), GDP increases by
approximately 24.8%., the amount of bank credit remaining constant.
•
In average, if FDI increases by 1 unit (1 million USD), GDP increases by
approximately 52.7%., the amount of bank credit remaining constant.
√
In average, if FDI increases by 1 unit (1 million USD) GDP increases by
approximately 22.2%., the amount of bank credit remaining constant.
170. In the following equation, GDP refers to gross domestic product, and FDI refers
the foreign direct investment. log(gdp) = 2.65 + 0.527*log(bankcredit) +
0.222*FDI. Which of the following statements is then true?
•
In average, if GDP increases by 1%., bank credit increases by log(0.527)%., the
level of FDI remaining constant.
•
In average, if GDP increases by 1%., bank credit increases by 0.527%., the level of
FDI remaining constant.
√
In average, if bank credit increases by 1%., GDP increases by 0.527%., the level of
FDI remaining constant.
•
In average, if bank credit increases by 1%., GDP increases by log(0.527)%., the
level of FDI remaining constant.
171. F-test is used to:
•
Economic significance
•
Individual statistical significance
•
All of them
√
Joint significance
✓
to test multiple variables at once
✓
determine whether joint impact of several independent variables over the
dependent variable is significant or not
✓
to test multiple linear restrictions
✓
testing exclusion restrictions
✓
test equality of two variances, analysis of variance, regression analysis
172. T-test is used to:
•
Determining whether joint impact of several independent variables over the
dependent variable is significant or not
•
Determining whether there is multicollinearity problem in the estimated model
or not
√
Determining whether there is any significant relationship between an
independent and dependent variable
•
Determining whether parameters of the estimated model are unbiased or not
•
Determining how much percent of variation in the dependent variable is
explained by the independent variables
173. Which one of the following statements is true?
√
Seasonality is not an issue when using annual (yearly) time series observations.
•
A trending variable cannot be used as the dependent variable in multiple
regression analysis.
•
Like cross-sectional observations, we can assume that most time series
observations are independently distributed.
•
The OLS estimator in a time series regression is biased under the first three
Gauss-Markov assumptions.
•
None of them
174. If the estimated model parameter is less than true model parameter, then the
coefficient is:
•
None of them
✓
Underestimated
•
Unbiased
•
Overestimated
175. If the estimated model parameter is greater than true model parameter, then the
coefficient is:
•
None of them
•
Underestimated
•
Unbiased
✓
Overestimated
176. Suppose true model is y = b0 + b1*x1 + b2*x2 +u. If x2 is not included to the
estimated model, then the estimated model most likely has:
✓
omitted variable biasedness,
✓
misspecification of the model
177. Which one of the following statements is true about unbiasedness of regression
parameters?
✓
Linear in parameters
✓
Random sampling
✓
The values of the explanatory variables are not all the same
✓
The value of the explanatory variable must contain no information about the
mean of the unobserved factors.
✓
There can not be omission of relevant variable
178. In BLUE condition, what “L” denotes:
✓
Linear
179. Why we need to know variance of parameters in an empirical analysis?
✓
True model`in estimated model`lə üst üstə düşdüyünü bilmək üçün
180. Which one of the following statements is most likely true?
√
A model represents a population relationship.
•
A model shows the estimated parameters of the explanatory variables.
•
In an empirical paper, the estimation methods should be discussed before
specifying a model.
•
The methods for estimating a model are same as the model itself.
•
None of them
181. In case of omitted variable biasedness, an omitted variable is considered to be:
✓
correlated with x
✓
determinant of y
✓
important factor
✓
significant factor
182. In a multiple regression model, omitting variable is:
✓
excluding an relevant variable
✓
correlated with x
✓
determinant of y
✓
significant variable
✓
important variable or factor
183. In a multiple regression model, an irrelevant factor is considered to be:
✓
a factor that is not correlated with dependent and independent variable
✓
increases the variance of the model
✓
does not effect the unbiasedness of OLS estimators
184. In a multiple regression model, adding a lot of irrelevant factors:
✓
increases the variance of the model
✓
does not effect the unbiasedness of OLS estimators
✓
a factor that is not correlated with dependent and independent variable
185. Consider the given model, log(price) = b0 + b1*log(area) + b2*distance. Note that
price, area and distance denote price (measured in AZN), largeness (measured in
m2) and distance from nearest metro station (measured in km) of a flat in Baku. If
b2=0.045, and standard error of b1 is 0.172, comment on statistical significance
of the corresponding association. Note that, critical value of t-test at 5%
significance level is 1.96.
•
The impact of price over distance of a flat is not statistically significant at 5% level
of significance.
•
The impact of price over distance of a flat is statistically significant at 5% level of
significance.
•
The impact of distance over price of a flat is statistically significant at 5% level of
significance.
√
The impact of distance over price of a flat is not statistically significant at 5% level
of significance.
•
The impact of distance from metro station over area is statistically significant at
5% level of significance.
186. A researcher wants to analyse the impact of a gender (male, female) in her
research on "CEO compensation" and "ROE" (Return on equity) of a firm. How
many dummy variables should the researcher include?
✓
1
187. Sales of an ice cream company depends on seasons and several other
quantitative factors like marketing expenses etc. Since there are 4 seasons in a
year, how many dummy variables should be included at most in the regression
equation for sales determination of an ice cream company?
✓
3
188. In the following model consumption is determined by income and gender.
Consumption = β0 + δ0 Gender + β1Inc + u
The variable Gender takes a value of 1 if the person is female and the variable
‘Inc’ measures the income of the individual. Refer to the model above. If δ0>0,__.
✓
Female’s consumption is higher than male’s consumption
✓
Male’s consumption is lower than female’s consumption
189.
In the following model consumption is determined by income and gender.
Consumption = β0 + δ0 Gender + β1Inc + u
The variable Gender takes a value of 1 if the person is male and the variable ‘Inc’
measures the income of the individual. Refer to the model above. The benchmark
group in this model is _______.
✓
female
190. Which of the following will be described using a binary variable?
√
Whether it rained on a particular day or it did not
•
The concentration of dust particles in air
•
The volume of rainfall during a year
•
The percentage of humidity in air on a particular day
191. While testing statistical significance on the association between an independent
variable and dependent variable, the confidence level must be at least
•
99.00%
•
80.00%
•
70.00%
•
60.00%
√
90.00%
192. t-test is used to:
✓
to test single variable
✓
determine whether there is any significant relationship between independent
and dependent variable
✓
to test whether differences between two means is statistically significant
193. Consider the given model, log(LS) = b0 + b1*log(income) + b2*work_hour
+b3*age. Note that LS represents life satisfaction of an individual (changing
between 5 and 35), income represents monthly income, work_hour shows
average weekly working hours, and age shows age of the individual. If b2 = 0.015
and standard error of b1 is 0.048, comment on statistical significance of the
association. Note that t-critical value is 2.57 at 99% confidence level, 1.96 at 5%
significance level, and 1.67 at 90% significance level. According to hypothesis test
result:
✓
The impact of average weekly working hours (monthly income) over life
satisfaction of an individual is not statistically significant. (Heç bir səviyyədə)
Qeyd: Şərtdə ya b1-i verməlidir, ya da b2-ni amma burda ikisi də verilib(güman ki
səhvdir), ona görə də tam olaraq hansı variable olduğunu bilməsək də cavab belədir.
194. Consider the given model, log(price) = b0 + b1*log(area) + b2*distance. Note that
price, area and distance denote price (measured in AZN), largeness (measured in
m2) and distance from nearest metro station (measured in km) of a flat in Baku. If
b1=1.209, and standard error of b1 is 0.045, comment on statistical significance
of the corresponding association. Note that, critical value of t-test at 5%
significance level is 1.96.
✓
The impact of largeness over price of a flat in Baku is statistically significant at 5%
level of significance level
195. Greater t-statistic value (or t-ratio) means
•
Less likely to reject the null hypothesis
•
None of them
✓
More likely to reject the null hypothesis
•
More likely to reject the alternative hypothesis
196. Consider the given model, Income=192.2+131.7 *log(work_hour)+4.396 *Exper.
Note that income denote monthly salary, and work_hour denote average
working hours per week while Exper represent total job experience (measured in
years) of the individual. How much income is predicted to change when weekly
working hours increase 5% while job experience does not change?
•
850.7 AZN
•
658.5 AZN
✓
6.585 AZN
•
855.1 AZN
197. Consider the given model, Income=192.2+131.7 *log(work_hour)+4.396 *Exper.
Note that income denote monthly salary, and work_hour denote average
working hours per week while Exper represent total job experience (measured in
years) of the individual. How much income is predicted to change when weekly
working hours increase 3% and job experience increase 2 years?
•
403.89 AZN
✓
12.743 AZN
•
204.94 AZN
•
596.1 AZN
198. Consider the given model, Income=192.2+131.7 *log(work_hour)+4.396 *Exper.
Note that income denote monthly salary, and work_hour denote average
working hours per week while Exper represent total job experience (measured in
years) of the individual. In the model, 4.396 means?
•
In average, ceteris paribus, 1 %. increase in total job experience is predicted to
increase income by 4.396 AZN.
•
In average, ceteris paribus, 1 year increase in total job experience is predicted to
increase income by 4.396%..
•
In average, ceteris paribus, 1 year increase in total job experience is predicted to
increase income by 0.04396%..
√
In average, ceteris paribus, 1 year increase in total job experience is predicted to
increase income by 4.396 AZN.
199. Consider the given model, Income=192.2+131.7 *log(work_hour)+4.396 *Exper.
Note that income denote monthly salary, and work_hour denote average
working hours per week while Exper represent total job experience (measured in
years) of the individual. In the model, 131.7 means?
•
In average, ceteris paribus, 1 hour increase in weekly working hours is expected
to decrease income by 131.7 AZN.
•
In average, ceteris paribus, 1 %. increase in weekly working hours is expected to
increase income by 1.317 %..
√
In average, ceteris paribus, 1 %. increase in weekly working hours is expected to
increase income by 1.317 AZN.
•
In average, ceteris paribus, 1 hour increase in weekly working hours is expected
to increase income by 131.7 AZN.
200.
Consider the given model, log(cons)=1.2+0.60*log(inc)+0.026*age. Note that
cons denote consumption, and inc denote monthly income while age represent
age of the individual. In the model, 0.60 means?
•
In average, ceteris paribus, when income decreases 1 %., consumption is
expected to increase 0.60 %..
•
In average, ceteris paribus, when income increases 1 AZN, consumption is
expected to increase 0.60 AZN.
•
In average, ceteris paribus, when income increases 1 AZN, consumption is
expected to increase 0.60%..
√
In average, ceteris paribus, when income increases 1 %., consumption is expected
to increase 0.60 %..
201.
Consider the given model, log(cons)=1.2+0.60*log(inc)+0.026*age. Note that
cons denote consumption, and inc denote monthly income while age represent
age of the individual. In the model, 0.026 means?
•
In average, ceteris paribus, when age increases 1 year, consumption is expected
to decrease 2.6%.
√
In average, ceteris paribus, when age increases 1 year, consumption is expected
to increase 2.6%.
•
In average, ceteris paribus, when age increases 1 year, consumption is expected
to increase 0.026%.
•
In average, ceteris paribus, when age increases 1 year, consumption is expected
to decrease 0.026%.
202.
Consider the given model, log(cons)=0.6+0.70*log(inc)-0.0035*age. Note that
cons denote consumption, and inc denote monthly income while age represent
age of the individual. How much consumption is predicted to change when age
decreases 3 years and income increases 10% while holding other factors fixed?
•
8, 65%.
√
8, 05%.
•
7, 6105%.
•
0, 6805%.
203.
Consider the given model, log(cons)=0.63+0.75*log(inc)-0.0025*age. Note that
cons denote consumption, and inc denote monthly income while age represent
age of the individual. How much consumption is predicted to change when age
decreases 3 years while holding other factors fixed?
•
1, 3575%.
•
2, 1%.
√
0, 75%.
•
-1, 5%.
204.
Consider the given model, log(cons)=0.52+0.325*log(inc)-0.025*age. Note that
cons denote consumption, and inc denote monthly income while age represent
age of the individual. How much consumption is predicted to change when
income increases 3 years while holding other factors fixed?
√
0, 975%.
•
-1, 005%.
•
1, 495%.
•
1, 47%.
205.
Sona’s research topic is to study families’ income elasticity of consumption in
Baku city center. For this purpose, he conducts survey among randomly selected
700 families living in Baku city center. Families have reported their average family
income (inc), consumption spending (cons) amount for a month, and number of
family members (fm). Note that income and consumption is measured in AZN,
and number of family members is measured as number of individuals in each
family. To get more reliable results, Sona estimates two different models: once
did not include fm, once did. So, estimated models are:
log(cons)=0.78+0.5*log(inc), and log(cons)=0.84+0.52*log(inc)+0.00256*fm. How
income elasticity of consumption changes when Sona control for number of
family members?
√
Increases slightly
•
Does not change
•
Decreases slightly
•
Decreases substantially
206.
Mammad’s research topic is to study families’ income elasticity of consumption
in Baku city center. For this purpose, he conducts survey among randomly
selected 700 families living in Baku city center. Families have reported their
average family income (inc), consumption spending (cons) amount for a month,
and number of family members (fm). Consider given model of the research,
log(cons)=2.25+0.72*log(inc)+0.026*fm. How much consumption is predicted to
change if income remains the same, and number of family members increases 2
people?
•
0, 772%.
√
5, 2%.
•
4, 48%.
•
0, 052%.
207.
Mammad’s research topic is to study families’ income elasticity of consumption
in Baku city center. For this purpose, he conducts survey among randomly
selected 700 families living in Baku city center. Families have reported their
average family income (inc), consumption spending (cons) amount for a month,
and number of family members (fm). Consider given model of the research,
log(cons)=2.25+0.72*log(inc)+0.026*fm. How much consumption is predicted to
change if income decreases 3%, and number of family members increases 2
people?
•
-3, 04%.
•
2, 108%.
√
3, 04%.
•
-2, 108%.
208.
Sona’s research topic is to study families’ income elasticity of consumption in
Baku city center. For this purpose, he conducts survey among randomly selected
700 families living in Baku city center. Families have reported their average family
income (inc), consumption spending (cons) amount for a month, and number of
family members (fm). Consider given model of the research,
log(cons)=0.84+0.50*log(inc)+0.00256*fm. How much consumption is predicted
to change if income increase 5% while fm does not change?
•
Does not change
•
1, 5%.
•
0, 125%.
√
2, 5%.
209.
A predicted value of a dependent variable:
•
represents the difference between the expected value of the dependent variable
and its actual value.
√
represents the expected value of the dependent variable given particular values
for the explanatory variables.
•
is independent of explanatory variables and can be estimated on the basis of the
residual error term only.
•
is always equal to the actual value of the dependent variable.
210.
Which of the following correctly identifies a limitation of logarithmic
transformation of variables?
•
Taking log of a variable often expands its range which can cause inefficient
estimates.
•
Taking log of variables make OLS estimates more sensitive to extreme values in
comparison to variables taken in level.
√
Logarithmic transformations cannot be used if a variable takes on zero or
negative values.
•
Logarithmic transformations of variables are likely to lead to heteroskedasticity.
211.
Heteroskedasticity means that
•
homogeneity cannot be assumed automatically for the model.
√
the variance of the error term is not constant.
•
the observed units have different preferences.
•
agents are not all rational
212.
Which one is a heteroskedasticity test?
√
Breusch-Pagan test,
√
White test
√
ARCH
√
Glejser
213.
Consider the following regression model: yi = β0 + β1xi + ui. If the first four
Gauss-Markov assumptions hold true, and the error term contains
heteroskedasticity, then _____.
✓
Var(ui|xi) = σ2
•
Var(ui|xi) = 0
•
Var(ui|xi) = 1
•
Var(ui|xi) = σ
214.
While testing statistical significance on the association between an independent
variable and dependent variable, to reject null hypothesis, the confidence level
must be at least
•
99%.
•
80%.
•
50%.
√
90%.
215.
To test individual statistical significance, which test is used?
•
F-test
•
None of them
√
t-test
•
Both of them
216.
While testing for statistical significance of the association between an
independent variable and dependent variable, null hypothesis means
•
All of them.
•
Sample regression parameter equals zero
•
There is no relationship between independent and dependent variable in the
sample
√
Population regression parameter equals zero.
217.
What is the primary goal of hypothesis testing (t-test)?
•
To identify if there is omitted variable biasedness in the model
•
To identify if there is significant association between independent variables
•
All of them
√
To identify whether sample regression function parameters can be generalized to
population parameters.
218.
Type I and Type II error. Which one is more crucial?
√
Type I
•
Equally crucial. Both error should be minimized
•
Type II
•
Neither Type I nor Type II.
219.
Type I error happens when:
•
You fail to reject true null hypothesis
•
None of them
√
You reject true null hypothesis
•
You reject true alternative hypothesis
220.
Which of the following correctly identifies an advantage of using adjusted R2 (Rsquared) over R2?
√
The penalty of adding new independent variables is better understood through
adjusted R2 than R2.
•
The adjusted R2 can be calculated for models having logarithmic functions while
R2 cannot be calculated for such models.
•
Adjusted R2 corrects the bias in R2.
•
Adjusted R2 is easier to calculate than R2.
221.
Which of the following statements is true?
√
The F statistic is always nonnegative as SSR of restricted model is never smaller
than SSR of unrestricted model.
•
The F statistic is more flexible than the t statistic to test a hypothesis with a single
restriction.
•
If the calculated value of F statistic is higher than the critical value, we reject the
alternative hypothesis in favor of the null hypothesis.
•
Degrees of freedom of a restricted model is always less than the degrees of
freedom of an unrestricted model.
222.
Which of the following statements is true of hypothesis testing?
√
A restricted model will always have fewer parameters than its unrestricted
model.
•
The t test can be used to test multiple linear restrictions.
•
A test of single restriction is also referred to as a joint hypotheses test.
•
OLS estimates maximize the sum of squared residuals.
223.
Which of the following tools is used to test multiple linear restrictions?
•
Unit root test
√
F test
•
z test
•
t test
224.
The general t-statistic value can be calculated as:
•
Estimated value of the coefficient minus hypothesized value, divided by variance
of the coefficient.
√
Estimated value of the coefficient minus hypothesized value, divided by standard
error of the coefficient.
•
Estimated value of the coefficient minus hypothesized value
•
Hypothesized value divided by standard error of the coefficient
225.
The significance level of a test is:
•
The probability of rejecting the null hypothesis when it is false.
•
One minus the probability of rejecting the null hypothesis when it is false.
•
One minus the probability of rejecting the null hypothesis when it is true.
√
The probability of rejecting the null hypothesis when it is true.
226.
Consider the equation, Y = β1 + β2*X2 + u. A null hypothesis, H0: β2 = 0 states
that:
•
X2 has no effect on the expected value of β2
•
Y has no effect on the expected value of X2.
√
X2 has no effect on the expected value of Y.
•
β2 has no effect on the expected value of Y.
227.
Which of the following is a statistic that can be used to test hypotheses about a
single population parameter?
•
F statistic
•
Wald statistic
√
t statistic
•
χ2 statistic
228.
Consider the given model, log(LS) = b0 + b1*log(income) + b2*work_hour
+b3*age. Note that LS represents life satisfaction of an individual (changing
between 5 and 35), income represents monthly income, work_hour shows
average weekly working hours, and age shows age of the individual. Null
hypothesis is, H0: b2=b3=0. F-test value is 11.73, and probability of F-statistic is
0.0000. According to hypothesis test result:
•
Joint impact of average weekly working hours and age over life satisfaction of
individuals is not statistically significant.
•
Joint impact of average weekly working hours and age over life satisfaction of
individuals is statistically significant at 10%. level of significance
•
Joint impact of average weekly working hours and age over life satisfaction of
individuals is statistically significant at 5%. level of significance
√
Joint impact of average weekly working hours and age over life satisfaction of
individuals is statistically significant at 1%. level of significance
229.
Consider the given model, log(LS) = b0 + b1*log(income) + b2*work_hour
+b3*age. Note that LS represents life satisfaction of an individual (changing
between 5 and 35), income represents monthly income, work_hour shows
average weekly working hours, and age shows age of the individual. Null
hypothesis is, H0: b1=b3=0. F-test value is 2.17, and probability of F-statistic is
0.1138. According to hypothesis test result:
•
Joint impact of monthly income and age over life satisfaction of individuals is
statistically significant at 5%. level of significance
•
Joint impact of monthly income and age over life satisfaction of individuals is
statistically significant at 1%. level of significance
•
Joint impact of monthly income and age over life satisfaction of individuals is
statistically significant at 10%. level of significance
√
Joint impact of monthly income and age over life satisfaction of individuals is
statistically insignificant.
230.
Consider the given model, log(LS) = b0 + b1*log(income) + b2*work_hour
+b3*age. Note that LS represents life satisfaction of an individual (changing
between 5 and 35), income represents monthly income, work_hour shows
average weekly working hours, and age shows age of the individual. Null
hypothesis is, H0: b1=b2=b3=0. F-test value is 7.82, and probability of F-statistic is
0.0001. According to hypothesis test result:
•
Joint impact of monthly income and average weekly work hour over life
satisfaction of individuals is statistically significant at 1%. Level of significance
•
Joint impact of monthly income, average weekly work hour and age over life
satisfaction of individuals is not statistically significant.
√
coint impact of monthly income, average weekly work hour and age over life
satisfaction of individuals is statistically significant at 1%. level of significance
•
Joint impact of monthly income and age over life satisfaction of individuals is
statistically significant at 1%. level of significance
231.
Consider the given model, log(LS) = b0 + b1*log(income) + b2*work_hour
+b3*age. Note that LS represents life satisfaction of an individual (changing
between 5 and 35), income represents monthly income, work_hour shows
average weekly working hours, and age shows age of the individual. If estimated
coefficient of b3 is 0.003 and standard error of b1 is 0.0013, comment on
statistical significance of the association. Note that t-critical value is 2.57 at 99%
confidence level, 1.96 at 5% significance level, and 1.67 at 90% significance level.
According to hypothesis test result:
•
The impact of age over life satisfaction of an individual is statistically insignificant
•
The impact of age over life satisfaction of an individual is statistically significant at
1%. level of significance
•
The impact of age over life satisfaction of an individual is statistically significant at
10%. level of significance
√
The impact of age over life satisfaction of an individual is statistically significant at
5%. level of significance
232.
Consider the given model, log(LS) = b0 + b1*log(income) + b2*work_hour
+b3*age. Note that LS represents life satisfaction of an individual (changing
between 5 and 35), income represents monthly income, work_hour shows
average weekly working hours, and age shows age of the individual. If estimated
coefficient of b2 is 0.015 and standard error of b1 is 0.048, comment on
statistical significance of the association. Note that t-critical value is 2.57 at 99%
confidence level, 1.96 at 5% significance level, and 1.67 at 90% significance level.
According to hypothesis test result:
•
The impact of monthly income over life satisfaction of an individual is statistically
significant at 5%. level of significance
•
The impact of monthly income over life satisfaction of an individual is statistically
significant at 1%. level of significance
•
The impact of monthly income over life satisfaction of an individual is statistically
significant at 10%. level of significance
√
The impact of monthly income over life satisfaction of an individual is not
statistically significant.
233.
Consider a given model, log(price) = b0 +b1*distance +b2*room +u. Note that
price represents price of houses in Baku city, distance represents distance of each
house from the nearest metro station, and room show number of rooms of each
house. If estimated coefficient of b2 is 0.412 and standard error of b2 is 0.018,
comment on statistical significance of the association. Note that t-critical value is
2.57 at 99% confidence level, 1.96 at 5% significance level, and 1.67 at 90%
significance level. According to hypothesis test result:
•
The impact of number of rooms over price of the house is statistically
insignificant
√
The impact of number of rooms over price of the house is statistically significant
at 1%. level of significance
•
The impact of number of rooms over price of the house is statistically significant
at 10%. level of significance
•
The impact of number of rooms over price of the house is statistically significant
at 5%. level of significance
234.
Consider a given model, log(price) = b0 +b1*distance +b2*room +u. Note that
price represents price of houses in Baku city, distance represents distance of each
house from the nearest metro station, and room show number of rooms of each
house. If estimated coefficient of b1 is 0.012 and standard error of b1 is 0.018,
comment on statistical significance of the association. Note that t-critical value is
2.57 at 99% confidence level, 1.96 at 5% significance level, and 1.67 at 90%
significance level. According to hypothesis test result:
•
The impact of distance from nearest metro station over price of the house is
statistically significant at 1%. level of significance
•
The impact of distance from nearest metro station over price of the house is
statistically significant at 10%. level of significance
√
The impact of distance from nearest metro station over price of the house is
statistically insignificant.
•
The impact of distance from nearest metro station over price of the house is
statistically significant at 5%. level of significance
235.
Consider a given model, log(income) = b0 + b1*log(work_hour) + b2*Exper +u.
Note that income denote monthly salary, and work_hour denote average
working hours per week while Exper represent total job experience (measured in
years) of the individual. If b2=0.006, and standard error of b2 is 0.002, comment
on statistical significance of the corresponding association. Note that, critical
value of t-test at 5% significance level is 1.96.
√
The impact of job experience over monthly salary is statistically significant at 5%.
level of significance level
•
None of them.
•
Population parameter of Exper is zero.
•
The impact of job experience over monthly salary is statistically insignificant at
5%. level of significance level
236.
Consider a given model, log(income) = b0 + b1*log(work_hour) + b2*Exper +u.
Note that income denote monthly salary, and work_hour denote average
working hours per week while Exper represent total job experience (measured in
years) of the individual. If b1=0.209, and standard error of b1 is 0.037, comment
on statistical significance of the corresponding association. Note that, critical
value of t-test at 5% significance level is 1.96.
•
Population parameter of log(work_hour) is zero.
√
The impact of working hour over monthly salary is statistically significant at 5%.
level of significance level
•
None of them.
•
The impact of working hour over monthly salary is statistically insignificant at 5%.
level of significance level
237.
Calculate interpret confidence intervals for a slope parameter – b1 at 1% level of
significance. Note that critical value of t-statistic is 2.57, and standard error of
that coefficient is 0.358. If b1=2.085 and show the impact of entrance score over
GPA of students, comment on statistical significance of the relationship according
to calculated confidence intervals.
√
The impact of entrance score over GPA is statistically significant at 1%. level of
significance
•
Confidence interval include zero which means there is significant association
•
The impact of entrance score over GPA is not statistically significant at 1%. level
of significance
•
Confidence interval does not include zero which means there is no significant
association
238.
Calculate interpret confidence intervals for a slope parameter – b1, at 5% level of
significance. Note that critical value of t-statistic is 1.96, and standard error of
that coefficient is 0.06. If b1=1.05 and show the impact of entrance score over
GPA of students, comment on statistical significance of the relationship according
to calculated confidence intervals.
√
The impact of entrance score over GPA is statistically significant at 5% level of
significance
239.
In hypothesis testing, P-value of a coefficient represent
√
Exact percentage of Type I error
•
None of them
•
Exact percentage of Type II error
•
Economic significance of the association
240.
Economic significance of a regression parameter relies on
•
Large t-statistic value
•
None of them
√
Practical interpretation (consistence with theory) of the coefficient
•
Less standard error of the coefficient
241.
F-test is used most likely to test
•
Economic significance
•
Individual statistical significance
•
All of them
√
Joint significance
242.
While testing statistical significance on the association between an independent
variable and dependent variable, greater t-statistic value (or t-ratio) means
•
None of them
•
Less likely to reject the null hypothesis
•
More likely to reject the alternative hypothesis
√
More likely to reject the null hypothesis
243.
Consider the given model log(price) = 7.61 +0.913*log(area) + 0.0697
*log(room)– 0.032*distance. Note that “price” is the price of houses (measured
in AZN) in Baku, “area” shows the largeness of the houses (measured in m^2),
“room” represent number of rooms in the house, and “distance” displays how far
the house is from the nearest metro station (measured in km). If Ramsey-Reset
test score is 7.65 and p-value is 0.0000:
√
there is misspecification problem
√
omission of interaction term or quadratics
244.
Consider the given model log(price) = 11.64 – 0.917*log(area) + 0.215
*log(area)^2 – 0.029*distance. Note that “price” is the price of houses (measured
in AZN) in Baku, “area” shows the largeness of the houses (measured in m^2) and
“distance” displays how far the house is from the nearest metro station
(measured in km). In average, what is the price of a house if area = 55, and
distance = 4 km?
√
80975
245.
Consider the given model log(price) = 11.64 – 0.917*log(area) + 0.215
*log(area)^2 – 0.029*distance. Note that “price” is the price of houses (measured
in AZN) in Baku, “area” shows the largeness of the houses (measured in m^2) and
“distance” displays how far the house is from the nearest metro station
(measured in km). If the largeness of a house increases from 45 m2 to 46 m2,
how much the price will change?
•
2.55%
•
18.43%
•
-0.92%
√
0.72%
246.
Consider the given model log(price) = 11.64 – 0.917*log(area) + 0.215
*log(area)^2 – 0.029*distance. Note that “price” is the price of houses (measured
in AZN) in Baku, “area” shows the largeness of the houses (measured in m^2) and
“distance” displays how far the house is from the nearest metro station
(measured in km). What is the threshold level of largeness?
√
8.43
247.
Consider the given model log(price) = 11.64 – 0.917*log(area) + 0.215
*log(area)^2 – 0.029*distance. Note that “price” is the price of houses (measured
in AZN) in Baku, “area” shows the largeness of the houses (measured in m^2) and
“distance” displays how far the house is from the nearest metro station
(measured in km). Based on the given model, what you can say about the
relationship between largeness and house prices in Baku?
√
largeness has threshold value of 8.43
√
largeness has U-shaped graph
√
largeness has increasing marginal effect on house price
√
relationship between largeness and house price is quadratic
248.
Consider the given model log(price) = 11.64 – 0.917*log(area) + 0.215
*log(area)^2 – 0.029*distance. Note that “price” is the price of houses (measured
in AZN) in Baku, “area” shows the largeness of the houses (measured in m^2) and
“distance” displays how far the house is from the nearest metro station
(measured in km). According to the estimated model, which one of the following
statements is true?
√
largeness has threshold value of 8.43
√
largeness has U-shaped graph
√
largeness has increasing marginal effect on house price
√
relationship between largeness and house price is quadratic
√
ceteris paribus, in average, when distance increases 1 unit, house price decreases
by 2.9%
√
there is negative relationship between distance and house price
249.
If there is functional form misspecification problem in the estimated multiple
regression model:
√
Model parameters are biased
√
Model parameters do not represent the true relationship in the population
√
Omission of interaction term or quadratics
250.
If there is functional form misspecification problem in the estimated model:
•
Model parameters are unbiased
•
Model paramteres represent the true relationship in the population
•
Model must be re-estimated by dependent variable in natural logarithm
√
None of them are true
251. Consider a given model, log(LS) = b0 + b1*log(work_hour) +
b2*log(work_hour)*age + u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), work_hour shows average weekly
working hours, and age shows age of the individual. Note that b1 equals (-0.151)
and b2 equals 0.0008, and both coefficients are statistically significant at 5%
significance level. If age equals 50, how much life satisfaction will change if
average weekly working hours increases 1%?
•
Will decrease 0.249%.
•
Will decrease 11.10%.
•
Will decrease 3.849%.
√
Will decrease 0.111 %.
252. Consider a given model, log(LS) = b0 + b1*log(work_hour) +
b2*log(work_hour)*age + u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), work_hour shows average weekly
working hours, and age shows age of the individual. Note that b1 equals (-0.151)
and b2 equals 0.0008, and both coefficients are statistically significant at 5%
significance level. If age equals 35, how much life satisfaction will change if
average weekly working hours increases 1%?
√
Will decrease 0.123%.
•
Will decrease 12.3%.
•
Will decrease 0.151%.
•
Will decrease 2.649%.
253. Consider a given model, log(LS) = b0 + b1*log(work_hour) +
b2*log(work_hour)*age + u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), work_hour shows average weekly
working hours, and age shows age of the individual. Note that b1 equals (-0.151)
and b2 equals 0.0008, and both coefficients are statistically significant at 5%
significance level. How much life satisfaction will change if average weekly
working hours increases 1%?
•
Will decrease 0.151%.
√
Will be different at different values of age.
•
Will decrease 0.1502%.
•
Will decrease 0.1518%.
254. Consider a given model, log(LS) = b0 + b1*log(work_hour) +
b2*log(work_hour)*age + u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), work_hour shows average weekly
working hours, and age shows age of the individual. If b2 is statistically significant
at 5% significance level, this means
•
The impact of age of average weekly working hours is statistically significant at
5%. significance level
•
The impact of age over life satisfaction is statistically significant at 5%.
significance level
•
The impact of average weekly working hours over life satisfaction is statistically
significant at 5%. significance level
√
The impact of age over the relationship between average weekly working hours
and life satisfaction is statistically significant at 5%. significance level
255. Consider a given model, log(LS) = b0 + b1*log(work_hour) +
b2*log(work_hour)*age + u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), work_hour shows average weekly
working hours, and age shows age of the individual. If b2 equals 0.0008, this
means
•
In average, ceteris paribus, when age increases 1 year, life satisfaction of
individuals increases 0.0008%..
•
In average, ceteris paribus, when age increases 1 year, average weekly working
hours of individuals increases 0.0008%..
√
In average, ceteris paribus, when age increases 1 year, the impact of average
weekly working hours over life satisfaction increases 0.0008%..
•
In average, ceteris paribus, when average weekly working hours increases 1%.,
life satisfaction of individuals increases 0.0008%..
256. Consider a given model, log(LS) = b0 + b1*log(work_hour) +
b2*log(work_hour)*age + u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), work_hour shows average weekly
working hours, and age shows age of the individual. What b2 coefficient displays
here?
√
The impact of age over the relationship between average weekly working hours
and life satisfaction.
•
The impact of age over life satisfaction
•
The impact of average weekly working hours over life satisfaction
•
The impact of age of average weekly working hours
257. Consider a given model, log(LS)= b0 +b1*log(income) +b2*log(income)^2
+b3*work_hour +b4*age +u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), income represents monthly income,
work_hour shows average weekly working hours, and age shows age of the
individual. “^2” display the quadratics. If b1 = 3.426, and b2= - 0.243, and both
coefficients are statistically significant at 5% significance level, how much is the
threshold level, and marginal impact of monthly income over life satisfaction if
income equals 1200 AZN? Note than log(1200)=7.09.
•
Threshold level: 1152; Marginal impact: 1.703%.
•
Threshold level: 1327; Marginal impact: 1.703%.
•
Threshold level: 1327; Marginal impact: -0.019%.
√
Threshold level: 1152; Marginal impact: -0.019%.
258. Consider a given model, log(LS)= b0 +b1*log(income) +b2*log(income)^2
+b3*work_hour +b4*age +u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), income represents monthly income,
work_hour shows average weekly working hours, and age shows age of the
individual. “^2” display the quadratics. If b1 = 4.039, and b2= - 0.362, and both
coefficients are statistically significant at 5% significance level, how much is
marginal impact of monthly income over life satisfaction if income equals 350
AZN?
•
4, 24%.
•
-249.4 AZN
•
1, 918%.
√
-0, 202%.
259. Consider a given model, log(LS)= b0 +b1*log(income) +b2*log(income)^2
+b3*work_hour +b4*age +u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), income represents monthly income,
work_hour shows average weekly working hours, and age shows age of the
individual. “^2” display the quadratics. If b1=-14.74, and b2=1.069 while
probability of b1 is 0.1212, and probability of b2 is 0.1134, what can be said
about the impact of monthly income over life satisfaction?
•
The impact of income over life satisfaction display diminishing marginal returns
•
None of them.
√
There is no significant parabolic association
•
The impact of income over life satisfaction display increasing marginal returns
260. Consider a given model, log(LS)= b0 +b1*log(income) +b2*log(income)^2
+b3*work_hour +b4*age +u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), income represents monthly income,
work_hour shows average weekly working hours, and age shows age of the
individual. “^2” display the quadratics. If b1 = 4.039, and b2= - 0.362, and both
coefficient are statistically significant at 5% significance level, what can be said
about the impact of monthly income over life satisfaction?
•
The impact of income over life satisfaction display increasing marginal returns,
affects negatively up to 264 AZN, and positively after.
•
The impact of income over life satisfaction display increasing marginal returns,
affects positively up to 264 AZN, and negatively after.
√
The impact of income over life satisfaction display diminishing marginal returns,
affects positively up to 264 AZN, and negatively after.
•
The impact of income over life satisfaction display diminishing marginal returns,
affects negatively up to 264 AZN, and positively after.
261. Consider a given model, log(LS)= b0 +b1*log(income) +b2*log(income)^2
+b3*work_hour +b4*age +u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), income represents monthly income,
work_hour shows average weekly working hours, and age shows age of the
individual. “^2” display the quadratics. Suppose that b1<0, and b2>0, and both
are statistically significant at 1% significance level. Which one of the following is
true about the impact of monthly income over life satisfaction of individuals?
√
The impact of monthly income over life satisfaction varies across the level of
monthly income, with increasing marginal returns.
•
The impact of monthly income over life satisfaction is unidirectional, always
positive.
•
The impact of monthly income over life satisfaction is unidirectional, always
negative.
•
The impact of monthly income over life satisfaction varies across the level of
monthly income, with diminishing marginal returns.
262. Consider a given model, log(LS)= b0 +b1*log(income) +b2*log(income)^2
+b3*work_hour +b4*age +u. Note that LS represents life satisfaction of an
individual (changing between 5 and 35), income represents monthly income,
work_hour shows average weekly working hours, and age shows age of the
individual. “^2” display the quadratics. Suppose that b1>0, and b2<0, and both
are statistically significant at 5% significance level. Which one of the following is
true about the impact of monthly income over life satisfaction of individuals?
•
The impact of monthly income over life satisfaction varies across the level of
monthly income, with increasing marginal returns.
√
The impact of monthly income over life satisfaction varies across the level of
monthly income, with diminishing marginal returns.
•
The impact of monthly income over life satisfaction is unidirectional, always
negative.
•
The impact of monthly income over life satisfaction is unidirectional, always
positive.
263. Which statement best represent the motivation for the use of interaction terms
in regression models?
•
Inclusion of interaction terms solves functional form misspecification problem
and increase the error term.
•
Inclusion of interaction terms allow to take into account the marginal changes in
a unidirectional association
•
Inclusion of interaction terms allow to estimate whether marginal impact of an
independent variable over a dependent variable depends on the value of
corresponding independent variable or not.
√
Inclusion of interaction terms allow to estimate whether marginal impact of an
independent variable over a dependent variable depends on the value of another
independent variable or not.
264. Which statement best represent the motivation for the use of quadratics in
regression models?
•
Inclusion of quadratics solves functional form misspecification problem and
increase the error term.
•
Inclusion of quadratics allow to take into account the marginal changes in a
unidirectional association
•
Inclusion of quadratics allow to estimate whether marginal impact of an
independent variable over a dependent variable depends on the value of another
independent variable or not.
√
Inclusion of quadratics allow to estimate whether marginal impact of an
independent variable over a dependent variable depends on the value of
corresponding independent variable or not.
265. Which of the following models is used quite often to capture decreasing or
increasing marginal effects of a variable?
•
Models with logarithmic functions
•
Models with binary variables
√
Models with quadratic functions
•
Models with variables in level
266. Consider the given model log(price) = 7.61 +0.913*log(area) + 0.0697
*log(room)– 0.032*distance. Note that “price” is the price of houses (measured
in AZN) in Baku, “area” shows the largeness of the houses (measured in m^2),
“room” represent number of rooms in the house, and “distance” displays how far
the house is from the nearest metro station (measured in km). If White test score
is 8.56 and p-value is 0.0000:
✓
there is heteroskedasticity problem
267. Consider the given model, log(hourly_salary) = b0 + b1*Gender +b2*Age
+b3*Exper + u. Note that hourly_salary denote per-hour job earnings, age and
exper represent age and job experience of each individual (observation). Gender
is a dummy variable equals 1 if the person is female, and 0 if the person is male. If
b1 equals (-0.238), this means
√
Ceteris paribus, in average, per-hour job earning of females is 23.8%. less than
per-hour job earning of males.
•
Ceteris paribus, in average, per-hour job earning of females is 0.238 AZN less
than per-hour job earning of males.
•
Ceteris paribus, in average, one-unit increase in gender variable is predicted to
decrease hourly salary by 0.238 unit.
•
Ceteris paribus, in average, one-unit increase in gender variable is predicted to
decrease hourly salary by 23.8%..
268. Consider the given model, log(hourly_salary) = b0 + b1*Gender +b2*Age
+b3*Exper + u. Note that hourly_salary denote per-hour job earnings, age and
exper represent age and job experience of each individual (observation). Gender
is a dummy variable equals 1 if the person is female, and 0 if the person is male. If
b1 is negative, this means?
•
Ceteris paribus, in average, per-hour job earning of females is greater than perhour job earning of males.
•
Ceteris paribus, in average, one-unit increase in gender variable is predicted to
decrease hourly salary by b1*100%..
•
Ceteris paribus, in average, one-unit increase in gender variable is predicted to
decrease hourly salary by b1 unit.
√
Ceteris paribus, in average, per-hour job earning of females is less than per-hour
job earning of males.
269. Which one of the following statement is false?
√
A binary variable is a variable whose value changes with a change in the number
of observations.
•
None of them.
•
The dummy variable coefficient for a particular group represents the estimated
difference in intercepts between that group and the base group.
•
The multiple linear regression model with a binary dependent variable is called
the linear probability model.
270. Which one of the following statement is true?
√
The dummy variable coefficient for a particular group represents the estimated
difference in intercepts between that group and the base group.
•
A dummy variable trap arises when a single dummy variable describes a given
number of groups.
•
A binary variable is a variable whose value changes with a change in the number
of observations.
•
None of them.
271. Which of the following Gauss-Markov assumptions is violated by the linear
probability model?
•
The assumption that none of the independent variables are constants.
•
The assumption of zero conditional mean of the error term.
•
The assumption of no exact linear relationship among independent variables.
√
The assumption of constant variance (homoscedasticity) of the error term.
272. Consider the following regression equation: y = β0+β1x1+…βk xk+ u .In which of
the following cases, the dependent variable is binary?
√
y indicates whether a female is working or not.
•
y indicates the gross domestic product of a country
•
y indicates household consumption expenditure
•
y indicates the number of children in a family
273. In the following regression equation, y is a binary variable:
y= β0+β1x1+…βk xk+ u
In this case, the estimated slope coefficient, β1 measures
•
the predicted change in the probability of success when x1 decreases by one unit,
everything else remaining constant
√
the predicted change in the probability of success when x1 increases by one unit,
everything else remaining constant
•
the predicted change in the value of y when x1 decreases by one unit, everything
else remaining constant
•
the predicted change in the value of y when x1 increases by one unit, everything
else remaining constant
274. Which of the following is true of dependent variables?
•
A dependent variable cannot have a qualitative meaning.
•
A dependent variable cannot have more than 2 values
•
A dependent variable can only have a numerical value.
√
A dependent variable can be binary.
275. The income of an individual in Baku depends on his ethnicity and several other
factors which can be measured quantitatively. If there are 5 ethnic groups in
Baku, how many dummy variables at most can be included in the regression
equation for income determination in Baku?
•
1
•
6
•
5
√
4
276. The following simple model is used to determine the annual savings of an
individual on the basis of his annual income and education.
Savings = β0+∂0 Edu + β1Inc+u
The variable ‘Edu’ takes a value of 1 if the person is educated and the variable
‘Inc’ measures the income of the individual. Refer to the above model. If ∂0 > 0,
this means:
•
uneducated people have higher savings than those who are educated
•
individual with lower income have higher savings
√
educated people have higher savings than those who are not educated
•
individuals with lower income have higher savings
277. The following simple model is used to determine the annual savings of an
individual on the basis of his annual income and education.
Savings = β0+∂0 Edu + β1Inc+u
The variable ‘Edu’ takes a value of 1 if the person is educated and the variable
‘Inc’ measures the income of the individual. The base (comparison) group in this
model is
•
the group of individuals with a high income
•
the group of individuals with a low income
•
the group of educated people
√
the group of uneducated people
278. The following simple model is used to determine the annual savings of an
individual on the basis of his annual income and education.
Savings = β0+∂0 Edu + β1Inc+u
The variable ‘Edu’ takes a value of 1 if the person is educated and the variable
‘Inc’ measures the income of the individual. The inclusion of another binary
variable in this model that takes a value of 1 if a person is uneducated, will give
rise to the problem of
•
omitted variable bias
•
heteroskedastcity
√
dummy variable trap
•
self-selection
279. Which of the following is true of dummy variables?
•
A dummy variable takes a value of 1 or 10.
•
A dummy variable always takes a value higher than 1.
•
A dummy variable always takes a value less than 1.
√
A dummy variable takes a value of 0 or 1.
280. In a regression model, which of the following will be described using a binary
variable?
√
Whether it rained on a particular day or it did not
•
The concentration of dust particles in air
•
The volume of rainfall during a year
•
The percentage of humidity in air on a particular day
281. A __ variable is used to incorporate qualitative information in a regression model.
•
dependent
•
continuous
√
dummy
•
predicted
282. Consider the given model Female_work = 1.11 – 0.067*log(husband_salary) +
0.003*age – 0.401 *School – 0.259* Collage + 0.04*Master + 0.213 *PhD. Note
that the dependent variable “Female_Work” is a dummy variable equals 1 if the
respondent (female) works, 0 otherwise. “Age” and “Husband_salary” denote the
respondent’s age and montly earnings of her husband. “School”, “Collage”,
“Master” and “PhD” are the dummy variable displaying the educational level of
the respondent, equals 1 for the corresponding level of education level, and gets
0 otherwise. Bachelor degree holders are left as the base (reference) group.
According to the estimated model, how much is the employment probability of a
female with master degree compared to a female with collage degree?
✓
0.299
283. Consider the given model Female_work = 1.11 – 0.067*log(husband_salary) +
0.003*age – 0.401 *School – 0.259* Collage + 0.04*Master + 0.213 *PhD. Note
that the dependent variable “Female_Work” is a dummy variable equals 1 if the
respondent (female) works, 0 otherwise. “Age” and “Husband_salary” denote the
respondent’s age and montly earnings of her husband. “School”, “Collage”,
“Master” and “PhD” are the dummy variable displaying the educational level of
the respondent, equals 1 for the corresponding level of education level, and gets
0 otherwise. Bachelor degree holders are left as the base (reference) group.
According to the estimated model, how much is the employment probability
difference between a female with school degree and a female with bachelor
degree?
✓
0.401
284. Consider the given model Female_work = 1.11 – 0.067*log(husband_salary) +
0.003*age – 0.401 *School – 0.259* Collage + 0.04*Master + 0.213 *PhD. Note
that the dependent variable “Female_Work” is a dummy variable equals 1 if the
respondent (female) works, 0 otherwise. “Age” and “Husband_salary” denote the
respondent’s age and montly earnings of her husband. “School”, “Collage”,
“Master” and “PhD” are the dummy variable displaying the educational level of
the respondent, equals 1 for the corresponding level of education level, and gets
0 otherwise. Bachelor degree holders are left as the base (reference) group. If pvalue of “– 0.067” is “0.0949”, and the p-value of “0.003” is “0.2571, which
statement is true?
✓
Age is not statistically significant
✓
Husband salary has statistically significance level of 10% (almost insignificant)
285. Consider the given model Female_work = 1.11 – 0.067*log(husband_salary) +
0.003*age – 0.401 *School – 0.259* Collage + 0.04*Master + 0.213 *PhD. Note
that the dependent variable “Female_Work” is a dummy variable equals 1 if the
respondent (female) works, 0 otherwise. “Age” and “Husband_salary” denote the
respondent’s age and montly earnings of her husband. “School”, “Collage”,
“Master” and “PhD” are the dummy variable displaying the educational level of
the respondent, equals 1 for the corresponding level of education level, and gets
0 otherwise. Bachelor degree holders are left as the base (reference) group.
Based on the estimated model, what can be said about the impact of educational
attainment?
✓
School and college female graduates’ probability to work is less than Bachelor
degree holders
✓
Master and PhD female graduates’ probability to work is greater than Bachelor
degree holders
✓
PhD holders have the highest probability to work
✓
School graduates have the lowest probability to work
286. Consider the given model Female_work = 1.11 – 0.067*log(husband_salary) +
0.003*age – 0.401 *School – 0.259* Collage + 0.04*Master + 0.213 *PhD. Note
that the dependent variable “Female_Work” is a dummy variable equals 1 if the
respondent (female) works, 0 otherwise. “Age” and “Husband_salary” denote the
respondent’s age and montly earnings of her husband. “School”, “Collage”,
“Master” and “PhD” are the dummy variable displaying the educational level of
the respondent, equals 1 for the corresponding level of education level, and gets
0 otherwise. Bachelor degree holders are left as the base (reference) group.
According to the given model, which statement is true?
•
Ceteris paribus, in average, 1% increase in husband’s salary decreases the
female’s probability to work by 0.067.
•
Ceteris paribus, in average, 1% increase in husband’s salary decreases the
number of employed females by 0.067.
•
Ceteris paribus, in average, 1% increase in husband’s salary decreases the
number of employed females by 0.00067.
✓
Ceteris paribus, in average, 1% increase in husband’s salary decreases the
female’s probability to work by 0.00067.
287. Consider the given model log(salary) = 4.89 + 0.005*age +0.314 *log(work_hour)
- 0.231*School - 0.314*Collage + 0.206 *Master + 0.467*PhD. Note that “salary”,
“age”, and “work_hour) represent monthly salary, age and weekly average
working hours of the respondent. “School”, “Collage”, “Master” and “PhD” are
the dummy variable displaying the educational level of the respondent, equals 1
for the corresponding level of education level, and gets 0 otherwise. Bachelor
degree holders are left as the base (reference) group. According to the given
model, how much is the salary gap between school graduates compared to
master graduates?
✓
43.7%
288. Consider the given model log(salary) = 4.89 + 0.005*age +0.314 *log(work_hour)
- 0.231*School - 0.314*Collage + 0.206 *Master + 0.467*PhD. Note that “salary”,
“age”, and “work_hour) represent monthly salary, age and weekly average
working hours of the respondent. “School”, “Collage”, “Master” and “PhD” are
the dummy variable displaying the educational level of the respondent, equals 1
for the corresponding level of education level, and gets 0 otherwise. Bachelor
degree holders are left as the base (reference) group. According to the given
model, how much is the salary gap between PhD graduates compared to master
graduates?
•
0.206%
•
20.6%
•
46.7%
✓
26.1%
289. Consider the given model log(salary) = 4.89 + 0.005*age +0.314 *log(work_hour)
- 0.231*School - 0.314*Collage + 0.206 *Master + 0.467*PhD. Note that “salary”,
“age”, and “work_hour) represent monthly salary, age and weekly average
working hours of the respondent. “School”, “Collage”, “Master” and “PhD” are
the dummy variable displaying the educational level of the respondent, equals 1
for the corresponding level of education level, and gets 0 otherwise. Bachelor
degree holders are left as the base (reference) group. According to the given
model, how much is the salary gap between collage graduates compared to
bachelor graduates?
✓
-31.4%
290. Consider the given model log(salary) = 4.89 + 0.005*age +0.314 *log(work_hour)
- 0.231*School - 0.314*Collage + 0.206 *Master + 0.467*PhD. Note that “salary”,
“age”, and “work_hour) represent monthly salary, age and weekly average
working hours of the respondent. “School”, “Collage”, “Master” and “PhD” are
the dummy variable displaying the educational level of the respondent, equals 1
for the corresponding level of education level, and gets 0 otherwise. Bachelor
degree holders are left as the base (reference) group. According to the given
model:
✓
School and college graduates earn less than Bachelor degree holders
✓
Master and PhD graduates earn more than Bachelor degree holders
✓
PhD holders are the highest earning group
✓
School graduates are the lowest earning group
✓
ceteris paribus, in average, 1 percent increase in work hour, increases salary by
0.314 percent
291. Consider the given model log(salary) = 5.65 + 0.009*age - 0.006*age*female +
0.122*log(work_hour) - 0.261 *female. Note that “salary”, “age” and
“work_hour” represents the respondent’s monthly salary, age and average
number of working hours in a week. “Female” is a dummy variable equals 1 if the
person is female, gets 0 otherwise. If Ramsey-Reset test score is 4.24 and p-value
is 0.0000, then:
✓
there is omitted interaction term or quadratics variable biasedness
✓
there is misspecification in the model
292. Consider the given model log(salary) = 5.65 + 0.009*age - 0.006*age*female +
0.122*log(work_hour) - 0.261 *female. Note that “salary”, “age” and
“work_hour” represents the respondent’s monthly salary, age and average
number of working hours in a week. “Female” is a dummy variable equals 1 if the
person is female, gets 0 otherwise. If the standard error of “-0.006” is “0.001”
(t=2.57 at 1% level), which one of the following statements is not true?
✓
Əslində “interaction term of age and female is significant” cavabda not true olanı
soruşursa, “interaction term of age and female is insignificant” seçin. Main point
budur, ancaq başqa cavablar da ola bilər. Yaş cinsi mənsubiyyətdən asılı olaraq
maaşa təsir edir. Əgər təsir etmir yazırsa that is not true and you should choose
that option
293. Consider the given model log(salary) = 5.65 + 0.009*age - 0.006*age*female +
0.122*log(work_hour) - 0.261 *female. Note that “salary”, “age” and
“work_hour” represents the respondent’s monthly salary, age and average
number of working hours in a week. “Female” is a dummy variable equals 1 if the
person is female, gets 0 otherwise. According to the estimated model, which one
of the following statements is true?
✓
ceteris paribus, in average 1 percent increase in working hours in a week,
increases monthly salary by 0.122 percent
✓
ceteris paribus, in average 1 unit increase in age, changes female’s monthly salary
by (0.9 - 0.6) 0.3 percent
✓
ceteris paribus, in average 1 unit increase in age, changes male’s monthly salary
by 0.9 percent
✓
ceteris paribus, in average, females’ monthly salary is less than male’s monthly
salary
294. Consider the given model log(salary) = 5.74 + 0.006*age + 0.124*log(work_hour)
-0.472 *female. Note that “salary”, “age” and “work_hour” represents the
respondent’s monthly salary, age and average number of working hours in a
week. “Female” is a dummy variable equals 1 if the person is female, gets 0
otherwise. According to the estimated model, which one of the following
statements is true?
✓
ceteris paribus, in average 1 percent increase in working hours in a week,
increases monthly salary by 0.124 percent
✓
ceteris paribus, in average 1 unit increase in age, changes monthly salary by 0.6
percent
✓
ceteris paribus, in average, females’ monthly salary is less than male’s monthly
salary by 47.2 percent
295. Consider the given model log(salary) = 5.74 + 0.006*age + 0.124*log(work_hour)
-0.472 *female. Note that “salary”, “age” and “work_hour” represents the
respondent’s monthly salary, age and average number of working hours in a
week. “Female” is a dummy variable equals 1 if the person is female, gets 0
otherwise. If standard error of “-0.472” is “0.022” (t=2.57 at 1% level), which one
of the following statements is true?
✓
ceteris paribus, in average, monthly salary of females is 47.2%. less than monthly
salary of males.
✓
female variable is statistically significant
296. Consider the given model, log(hourly_salary) = b0 + b1*School +b2*Collage
+b3*Master +b4*PhD +b5*Age +b6*Exper + u. Note that hourly_salary denote
per-hour job earnings, age and exper represent age and job experience of each
individual (observation). School is a dummy variable equals 1 if the person’s
highest educational attainment level is graduation from comprehensive schools
and 0 otherwise. Collage is a dummy variable equals 1 if the person’s highest
educational attainment level is graduation from collages or vocational schools
and 0 otherwise. Master is a dummy variable equals 1 if the person’s highest
educational attainment level is completion of master degree and 0 otherwise.
PhD is a dummy variable equals 1 if the person’s highest educational attainment
level is completion of PhD (Doctor of Philosophy) degree and 0 otherwise. Note
that those with bachelor degree are left as the base group. If b1 = - 0.369, b2 = 0.319, b3 = 0.213, and b4 = 0.450, and all coefficients are statistically significant
at 5% significance level, calculate per-hour earning difference between PhD
graduates and collage graduates.
•
Ceteris paribus, in average, PhD graduates earn 45%. more than collage
graduates.
•
Ceteris paribus, in average, PhD graduates earn 31.9%. more than collage
graduates.
•
Ceteris paribus, in average, PhD graduates earn 13.1%. more than collage
graduates.
√
Ceteris paribus, in average, PhD graduates earn 76.9%. more than collage
graduates.
297. Consider the given model, log(hourly_salary) = b0 + b1*School +b2*Collage
+b3*Master +b4*PhD +b5*Age +b6*Exper + u. Note that hourly_salary denote
per-hour job earnings, age and exper represent age and job experience of each
individual (observation). School is a dummy variable equals 1 if the person’s
highest educational attainment level is graduation from comprehensive schools
and 0 otherwise. Collage is a dummy variable equals 1 if the person’s highest
educational attainment level is graduation from collages or vocational schools
and 0 otherwise. Master is a dummy variable equals 1 if the person’s highest
educational attainment level is completion of master degree and 0 otherwise.
PhD is a dummy variable equals 1 if the person’s highest educational attainment
level is completion of PhD (Doctor of Philosophy) degree and 0 otherwise. Note
that those with bachelor degree are left as the base group. If b1 = - 0.369, b2 = 0.319, b3 = 0.213, and b4 = 0.450, and all coefficients are statistically significant
at 5% significance level, calculate per-hour earning difference between master
graduates and school graduates.
√
Ceteris paribus, in average, master graduates earn 58.2%. more than school
graduates.
•
Ceteris paribus, in average, master graduates earn 15.6%. more than school
graduates.
•
Ceteris paribus, in average, master graduates earn 36.9%. more than school
graduates.
•
Ceteris paribus, in average, master graduates earn 21.3%. more than school
graduates.
298. Consider the given model, log(hourly_salary) = b0 + b1*School +b2*Collage
+b3*Master +b4*PhD +b5*Age +b6*Exper + u. Note that hourly_salary denote
per-hour job earnings, age and exper represent age and job experience of each
individual (observation). School is a dummy variable equals 1 if the person’s
highest educational attainment level is graduation from comprehensive schools
and 0 otherwise. Collage is a dummy variable equals 1 if the person’s highest
educational attainment level is graduation from collages or vocational schools
and 0 otherwise. Master is a dummy variable equals 1 if the person’s highest
educational attainment level is completion of master degree and 0 otherwise.
PhD is a dummy variable equals 1 if the person’s highest educational attainment
level is completion of PhD (Doctor of Philosophy) degree and 0 otherwise. Note
that those with bachelor degree are left as the base group. If b1 is negative and
statistically significant at 5% significance level, this means:
•
Ceteris paribus, in average, school graduates earn significantly more than others
with higher educational attainment level.
√
Ceteris paribus, in average, school graduates earn b1*100%. for per-hour less
than bachelor degree holders.
•
Ceteris paribus, in average, school graduates earn b1 AZN for per-hour less than
bachelor degree holders.
•
Ceteris paribus, in average, 1 unit change in school variable has statistically
significant impact over per-hour job earnings.
299. Consider the given model, log(hourly_salary) = b0 + b1*Gender +b2*Age
+b3*Exper + u. Note that hourly_salary denote per-hour job earnings, age and
exper represent age and job experience of each individual (observation). Gender
is a dummy variable equals 1 if the person is female, and 0 if the person is male. If
probability of b1 is greater than 10%, this means
•
Ceteris paribus, in average, per-hour job earning of females is significantly more
than per-hour job earning of males.
•
None of them.
√
There is no significant difference in per-hour job earning among males and
females
•
Ceteris paribus, in average, per-hour job earning of females is significantly less
than per-hour job earning of males.
300. Consider the given model, Female_work = b0 + b1*age + b2*log(salary) +u. Note
that Female_work is a dummy variable equals 1 if a married female works, 0 if
the married female does not work. Age represents age of the married female,
and salary shows average monthly earnings of female’s husband. b0=17.2,
b1=0.012 and b2= - 2.52, and both coefficients are statistically significant at 5%
significance level. What is predicted probability of a female to work if age equals
30, and husband’s salary is 800 AZN?
•
0.225
•
0.652
•
0.84
√
0.715
301. Consider the given model, Female_work = b0 + b1*age + b2*log(salary) +u. Note
that Female_work is a dummy variable equals 1 if a married female works, 0 if
the married female does not work. Age represents age of the married female,
and salary shows average monthly earnings of female’s husband. b0=15.6,
b1=0.012 and b2= - 2.52, and both coefficients are statistically significant at 5%
significance level. What is predicted probability of a female to work if age equals
35, and husband’s salary is 500 AZN?
•
0, 463
√
0, 359
•
0, 65
•
0, 24
302. Consider the given model, Female_work = b0 + b1*age + b2*log(salary) +u. Note
that Female_work is a dummy variable equals 1 if a married female works, 0 if
the married female does not work. Age represents age of the married female,
and salary shows average monthly earnings of female’s husband. If b2 = - 20.52
and statistically significant at 5% significance level, this means
•
Ceteris paribus, in average, when husbands’ salary increases one %., probability
of a married female to work decreases 20.52 points.
•
Ceteris paribus, in average, when husbands’ salary increases one AZN, number of
working females will decrease 20.52 unit
•
Ceteris paribus, in average, when husbands’ salary increases one 1%.,
female_work variable is predicted to decrease 20.52 points.
✓
Ceteris paribus, in average, when husbands’ salary increases one %., probability
of a married female to work decreases 0.2052 points.
303. Consider the given model, Female_work = b0 + b1*age + b2*log(salary) +u. Note
that Female_work is a dummy variable equals 1 if a married female works, 0 if
the married female does not work. Age represents age of the married female,
and salary shows average monthly earnings of female’s husband. If b1 = 0.012
and statistically significant at 5% significance level, this means
•
Ceteris paribus, in average, when age increases one year, probability of a married
female to work increases 0.012%..
•
Ceteris paribus, in average, when age increases one year, female_work variable is
predicted to increase 0.012 points.
•
Ceteris paribus, in average, when age increases one year, number of working
females will increase 1.2%.
✓
Ceteris paribus, in average, when age increases one year, probability of a married
female to work increases 0.012 points.
304. What is the main characteristics of linear probability models?
•
Dependent variable must be in natural logarithmic form
•
There can be limited number of independent variables
•
Dependent variable must be linear does not matter it is binary or quantitative
(scale) variable.
√
Dependent variable must be binary variable.
305. Consider the given model, log(Cons)=b0 +b1*log(income)
+b2*log(income)*Gender. Note that cons represent average monthly
consumption while income represent average monthly income of an individual.
Both variables are measured in AZN. Gender is a dummy variable equals 1 if the
person is female, and 0 if the person is male. Note that b1=0.56, and b2 = 0.24,
and both coefficients are statistically significant at 5% significance level.
According to the model, how much consumption of females changes if income
increases 1%, ceteris paribus, in average
•
0, 56%.
•
0, 24%.
√
0, 8%.
•
24%.
306. Consider the given model, log(Cons)=b0 +b1*log(income)
+b2*log(income)*Gender. Note that cons represent average monthly
consumption while income represent average monthly income of an individual.
Both variables are measured in AZN. Gender is a dummy variable equals 1 if the
person is female, and 0 if the person is male. Note that b1=0.56, and b2 = 0.24,
and both coefficients are statistically significant at 5% significance level.
According to the model, how much consumption of males changes if income
increases 1%, ceteris paribus, in average?
√
0, 56%.
•
24%.
•
0, 8%.
•
0, 24%.
307. Consider the given model, log(Cons)=b0 +b1*log(income)
+b2*log(income)*Gender. Note that cons represent average monthly
consumption while income represent average monthly income of an individual.
Both variables are measured in AZN. Gender is a dummy variable equals 1 if the
person is female, and 0 if the person is male. If b2 is positive and statistically
significant at 5% significance level, this means
•
Ceteris paribus, in average, 1%. increase in income of females increases
consumption of females b2%..
•
Ceteris paribus, in average, 1%. increase in income increases consumption b2%.
•
Ceteris paribus, in average, 1%. increase in income of males increases
consumption of males b2%..
√
Income elasticity of consumption of females is b2%. higher than income elasticity
of consumption of males
308. Consider the given model, log(Cons)=b0 +b1*log(income)
+b2*log(income)*Gender. Note that cons represent average monthly
consumption while income represent average monthly income of an individual.
Both variables are measured in AZN. Gender is a dummy variable equals 1 if the
person is female, and 0 if the person is male. Here, b1 means that
√
Ceteris paribus, in average, 1%. increase in income of males increases
consumption of males b1%..
•
Ceteris paribus, in average, 1%. increase in income increases consumption b1%.
•
Ceteris paribus, in average, 1%. increase in income of females increases
consumption of females b1%..
•
Income elasticity of consumption of females is b1%. higher than income elasticity
of consumption of males.
309. Consider the given model, log(Cons)=b0 +b1*log(income)
+b2*log(income)*Gender. Note that cons represent average monthly
consumption while income represent average monthly income of an individual.
Both variables are measured in AZN. Gender is a dummy variable equals 1 if the
person is female, and 0 if the person is male. If b2 is positive and statistically
significant at 5% significance level, this means
•
Income elasticity of consumption is higher for males than females.
•
Females’ income is higher than males’ income.
√
Income elasticity of consumption is lower for males than females.
•
Females spend more than females while assuming income to be the same.
310. Consider the given model, log(hourly_salary) = b0 + b1*School +b2*Collage
+b3*Master +b4*PhD +b5*Age +b6*Exper + u. Note that hourly_salary denote
per-hour job earnings, age and exper represent age and job experience of each
individual (observation). School is a dummy variable equals 1 if the person’s
highest educational attainment level is graduation from comprehensive schools
and 0 otherwise. Collage is a dummy variable equals 1 if the person’s highest
educational attainment level is graduation from collages or vocational schools
and 0 otherwise. Master is a dummy variable equals 1 if the person’s highest
educational attainment level is completion of master degree and 0 otherwise.
PhD is a dummy variable equals 1 if the person’s highest educational attainment
level is completion of PhD (Doctor of Philosophy) degree and 0 otherwise. Note
that those with bachelor degree are left as the base group. If b1 = - 0.369, b2 = 0.319, b3 = 0.213, and b4 = 0.450, and all coefficients are statistically significant
at 5% significance level, calculate per-hour earning difference between collage
graduates and school graduates.
•
Ceteris paribus, in average, collage graduates earn 36.9%. more than school
graduates
•
Ceteris paribus, in average, school graduates earn 5%. more than collage
graduates
√
Ceteris paribus, in average, collage graduates earn 5%. more than school
graduates
•
Ceteris paribus, in average, collage graduates earn 31.9%. more than school graduates
311. “OPRC” represents Azerbaijan’s average quarterly oil price during 2000Q1 –
2018Q4. To examine whether OPRC is stationary or non-stationary, ADF test
applied (with only intercept). If ADF test score at level is “-1.32” and p-value is
“0.6043”, while ADF score at first difference is 4.36 and p-value is 0.0345, then:
✓
OPRC is non-stationary at level
✓
OPRC is stationary at first difference
312. “OPRC” represents Azerbaijan’s average quarterly oil price during 2000Q1 –
2018Q4. To examine whether OPRC is stationary or non- stationary, ADF test
applied (with trend and intercept and at level). If ADF test score is “-5.45” and pvalue is “0.0000”, then:
✓
OPRC is trend stationary at level
313. “BE” represents budget expenditures of Azerbaijan on quarterly basis during
2000Q1 – 2018Q4. To examine whether BE is stationary or non-stationary, ADF
test applied (with only intercept and at level). If ADF test score is “-1.34” and pvalue is “0.6073”, then:
✓
BE is non-stationary at level
314. “GDP” represents Gross Domestic Product of Azerbaijan on quarterly basis during
2000Q1 – 2018Q4. To examine whether GDP is stationary or non-stationary, ADF
test applied (with only intercept and at level). If ADF test score is “-0.514” and pvalue is “0.8813”, then:
✓
GDP is non-stationary at level
315. If Augmented Dickey-Fuller test result in large p-value (greater than 10%) with
intercept at level and low p-value (less than 5%) with trend and intercept at level,
this means:
•
The variable is non-stationary at level, stationary at first difference
•
The variable is non-stationary at level, and at first difference
✓
The variable is trend-stationary at level.
•
The variable is stationary at level, and at first difference
316. If Augmented Dickey-Fuller test result in large p-value (greater than 10%) at level
and low p-value (less than 5%) at first difference, this means:
•
The variable is stationary at level, and at first difference
•
The variable is stationary at level, non-stationary at first difference
•
The variable is non-stationary at level, and at first difference
✓
The variable is non-stationary at level, stationary at first difference
317. If Kwiatkowski–Phillips–Schmidt–Shin (KPSS) test result in LM-statistic value less
than the critical value at 10% significance level:
√
The variable is stationary
318. If Kwiatkowski–Phillips–Schmidt–Shin (KPSS) test result in LM-statistic value
greater than the critical value at 5% significance level:
•
The variable has no unit root problem
√
The variable is non-stationary
•
The variable has normal distribution
•
The variable is stationary
319. Null hypothesis in Kwiatkowski–Phillips–Schmidt–Shin (KPSS) test is:
✓
The variable is stationary
•
The variable has unit root
•
The variable has normal distribution
•
The variable is non-stationary
320. Null hypothesis in Augmented Dickey-Fuller test is:
√
The variable is non-stationary
•
The variable has normal distribution
•
The variable has is stationary
•
The variable has no unit root
321. Which of the following test is used to identify existence of unit root problem in
variables?
✓
Augmented Dickey Fuller test
•
White test
•
Johansen test
•
Breusch-Pagan-Godfrey test
322. Which of the following is a test for serial correlation in the error terms?
•
Johansen test
•
White test
✓
Breusch-Pagan-Godfrey test
•
Augmented Dickey Fuller test
323. Which of the following is assumed in time series regression?
•
The error terms are contemporaneously heteroskedastic.
•
The explanatory variables are contemporaneously endogenous.
•
The explanatory variables cannot have temporal ordering.
✓
There is no perfect collinearity between the explanatory variables.
324. A process is stationary if:
✓
any collection of random variables in a sequence is taken and shifted ahead by h
time periods, the joint probability distribution remains unchanged (mean value
and variance do not change).
•
any collection of random variables in a sequence is taken and shifted ahead by h
time periods; the joint probability distribution changes.
•
there is serial correlation between the error terms of successive time periods and
the explanatory variables and the error terms have positive covariance.
•
there is no serial correlation between the error terms of successive time periods
and the explanatory variables and the error terms have positive covariance.
325. Which one of the following statements is true?
•
Economic time series are not outcomes of random variables.
•
In a static model, one or more explanatory variables affect the dependent
variable with a lag.
✓
Dummy variables can be used to address the problem of seasonality in regression
models.
•
Time series regression is based on series which exhibit serial correlation.
326.
A seasonally adjusted series is one which:
•
has had seasonal factors added to it.
•
has qualitative dependent variables representing different seasons.
✓
has seasonal factors removed from it.
•
has qualitative explanatory variables representing different seasons.
327. Which of the following is an assumption on which time-series regression is
based?
•
A time series process follows a model that is nonlinear in parameters.
•
For each time period, the expected value of the error ut, given the explanatory
variables for all time periods, is positive.
•
In a time-series process, at least one independent variable is a constant.
✓
In a time-series process, no independent variable is a perfect linear combination
of the others.
328. A static model is postulated when:
•
a change in the independent variable at time ‘t’ does not have any effect on the
dependent variable.
•
a change in the independent variable at time ‘t’ is believed to have an effect on
the dependent variable at period ‘t + 1’.
•
a change in the independent variable at time ‘t’ is believed to have an effect on
the dependent variable for all successive time periods.
✓
a change in the independent variable at time ‘t’ is believed to have an immediate
effect on the dependent variable.
329. Which of the following correctly identifies a difference between cross-sectional
data and time series data?
•
Cross-sectional data consists of only qualitative variables, whereas time series
data consists of only quantitative variables.
•
Cross-sectional data is based on temporal ordering, whereas time series data is
not.
•
Time series data consists of only qualitative variables, whereas cross-sectional
data does not include qualitative variables.
✓
Time series data is based on temporal ordering, whereas cross-sectional data is
not.
330. What will you conclude about a time-series regression model if the BreuschPagan-Godfrey test results in a small p-value (less than 5%)?
√
The model has heteroscedasticity problem.
•
The model has homoscedasticity problem.
•
The model omits some important explanatory factors.
•
The model contains dummy variables.
331. What will you conclude about a time-series regression model if the BreuschPagan-Godfrey test results in a large p-value (greater than 10%)?
•
The model has heteroscedasticity problem.
•
The model omits some important explanatory factors.
✓
The model has no heteroscedasticity problem.
•
The model contains dummy variables.
332. Which of the following tests can be used to test for heteroscedasticity in a timeseries?
•
Breusch-Pagan-Gordfrey test
•
Dickey-Fuller test
√
Johansen test
•
Durbin’s alternative test
333. In the presence of serial correlation:
•
estimated standard errors remain valid.
•
estimated test statistics remain valid.
✓
estimated OLS values are not BLUE.
•
estimated variance does not differ from the case of no serial correlation.
334. Which of the following is true of Regression Specification Error Test (RESET)?
•
It helps in the detection of multicollinearity among the independent variables in a
regression model.
•
It detects the presence of dummy variables in a regression model.
•
It helps in the detection of heteroscedasticity when the functional form of the
model is correctly specified.
✓
It tests if the functional form of a regression model is misspecified.
335. Which of the following is true?
•
A functional form misspecification occurs only if a key variable is uncorrelated
with the error term.
✓
A functional form misspecification can occur if the unidirectional impact of an
independent variable is estimated while there is quadratic association.
•
A functional form misspecification does not lead to inconsistency in the ordinary
least squares estimators.
•
A functional form misspecification does not lead to biasedness in the ordinary
least squares estimators.
336. A regression model suffers from functional form misspecification if.
•
the dependent variable is binary.
•
a key variable is binary
•
the coefficient of a key variable is zero.
√
an interaction term is omitted.
337. Which of the following is true of the White test?
•
The White test cannot detect forms of heteroscedasticity that invalidate the
usual Ordinary Least squares standard errors.
•
The White test is used to detect the presence of multicollinearity in a linear
regression model.
•
The White test can detect the presence of heteroscedasticity in a linear
regression model even if the functional form is misspecified.
✓
The White test assumes that the square of the error term in a regression model is
uncorrelated with all the independent variables, their squares and cross products
338. Which of the following is a difference between the White test and the BreuschPagan test?
•
The number of regressors used in the Breusch-Pagan test is larger than the
number of regressors used in the White test.
•
The White test is used for detecting heteroscedasticity in a linear regression
model while the Breusch-Pagan test is used for detecting autocorrelation.
•
The White test is used for detecting autocorrelation in a linear regression model
while the Breusch-Pagan test is used for detecting heteroscedasticity.
✓
The number of regressors used in the White test is larger than the number of
regressors used in the Breusch-Pagan test.
339. What will you conclude about a regression model if the White test results in a
large p-value (greater than 10%)?
✓
The model has no heteroscedasticity problem.
•
The model has heteroscedasticity problem.
•
The model contains dummy variables.
•
The model omits some important explanatory factors.
340. Consider the given model Log(GDP) = 3.46 + 0.755* log(BE) – 0.229* log(OPRC) –
0.0002*OPRN. Note that GDP, BE, OPRC, and OPRN represent Gross Domestic
Product, Budget Expenditure, oil price and oil production, respectively. In this
model, if ARCH Test score is 0.603 and its p-value is 0.6628, then in the model:
✓
In this model there is no heteroskedasticity problem
341. Consider the given model Log(GDP) = 3.46 + 0.755* log(BE) – 0.229* log(OPRC) –
0.0002*OPRN. Note that GDP, BE, OPRC, and OPRN represent Gross Domestic
Product, Budget Expenditure, oil price and oil production, respectively. In this
model, if White Test score is 1.21 and its p-value is 0.3018, then in the model:
✓
In this model there is no heteroskedasticity problem
342. Consider the given model Log(GDP) = 3.46 + 0.755* log(BE) – 0.229* log(OPRC) –
0.0002*OPRN. Note that GDP, BE, OPRC, and OPRN represent Gross Domestic
Product, Budget Expenditure, oil price and oil production, respectively. In this
model, if Breusch- Godfrey LM Test score is 8.36 and its p-value is 0.0000, then in
the model:
✓
In this model there is heteroskedasticity problem
343. What will you conclude about a regression model if the Breusch-Pagan test
results in a large p-value (greater than 10%)?
•
The model contains dummy variables.
•
The model has heteroscedasticity problem.
•
The model omits some important explanatory factors.
√
The model has no heteroscedasticity problem.
344. What will you conclude about a regression model if the White test results in a
small p-value (less than 5%)?
•
The model has homoscedasticity problem.
•
The model omits some important explanatory factors.
✓
The model has heteroscedasticity problem.
•
The model contains dummy variables.
345. What will you conclude about a regression model if the Breusch-Pagan test
results in a small p-value (less than 5%)?
✓
The model has heteroscedasticity problem.
•
The model has homoscedasticity problem.
•
The model contains dummy variables.
•
The model omits some important explanatory factors
346. Which of the following tests helps in the detection of heteroscedasticity?
•
T-test
•
The Chow test
✓
The Breusch-Pagan-Godfrey test
•
The Durbin-Watson test
347. Which of the following is true of heteroscedasticity?
•
Heteroscedasticity causes inconsistency in the Ordinary Least Squares estimators.
•
It is not possible to obtain F statistics that are robust to heteroscedasticity of an
unknown form.
✓
The Ordinary Least Square estimators are not the best linear unbiased estimators
if heteroscedasticity problem exists.
•
Population R2 (goodness of fit) is affected by the presence of heteroscedasticity.
348. Which of the following statements is true of an appropriate data set?
•
An appropriate data set should not be based on surveys.
•
An appropriate data set should be collected from government registered
websites.
•
An appropriate data set should not have time series units.
✓
An appropriate data set should have enough controls to do a reasonable ceteris
paribus analysis.
349. During the data collection process for an empirical research project, time series
data should be stored:
•
with the earliest time period listed as the last observation, and the most recent
time period as the first observation
✓
with the earliest time period listed as the first observation, and the most recent
time period as the last observation.
•
with the time period in which the concerned variable takes the highest value
listed as the first observation, and the time period in which the concerned
variable takes the lowest value as the last observation.
•
with the time period in which the concerned variable takes the lowest value
listed as the first observation, and the time period in which the concerned
variable takes the highest value as the last observation.
350. Suppose that you want to examine whether income elasticity of consumption is
different for males and females while controlling for age, education level and
marital status of respondents. In the model, you should include:
✓
An interaction term of log(income) with a dummy variable representing gender
status
351. If there is functional form misspecification problem according to Ramsey-Reset
test result, most likely you should
•
Take log of the dependent variable
•
Use binary dependent variable
√
Include quadratics or interaction terms of independent variables
•
Collect additional data and increase your sample size
352. If your outcome (dependent) variable is a dummy variable, you should estimate
353. When you use time-series data in an empirical research, estimation of
econometric model starts with
√
Testing for stationarity of all model variables
•
Testing for heteroscedasticity in the model
•
Testing for serial-correlation
•
Testing for functional form misspecification
•
None of them
354. While doing an empirical research by using cross sectional data, you do not need
to test for
•
Existence of heteroscedasticity
•
All of them
√
Existence of serial correlation
•
Existence of functional form misspecification
355. Which one of the following statements is incorrect about empirical research
process?
•
Collected data is used to estimate model parameters
•
Data collection stage comes after mathematical model specification
•
Mathematical specification of econometric model should refer to a theory
√
Estimated econometric model can be interpreted without application of any tests
356. An empirical research process starts with
•
Hypothesis testing
•
Specification of econometric model
•
Data collection
√
Statement of theory or hypothesis
357. Suppose you would like to test validity of Philips curve (association between
inflation and unemployment) among 17 developed countries within a single
empirical analyses framework. Which type of data you will most likely use?
√
Panel data
•
Pool data
•
Cross sectional data
•
Time series data
358. Suppose you would like to test validity of Philips curve (association between
inflation and unemployment) in Azerbaijan. Which type of data you will most
likely use?
•
Panel data
•
Cross sectional data
•
Pool data
√
Time series data
359. Suppose that you want to investigate income elasticity of consumption among
citizens of Azerbaijan Republic in 2019. Which type of data you should most likely
use?
•
Time series data
•
Panel data
√
Cross sectional data
•
Pool data