Uploaded by boveriellie

310quizlet

advertisement
BUAD 310: Final Exam
Study online at https://quizlet.com/_5rrk1n
1. Correlation
What does r in between
(-1 - 1) mean?
Difference between Correlation and Causation
A measure of the relationship between two variables
-1= Strong Negative Correlation
0= No relationship between x and y
+1= Strong Positive Correlation
Correlation doesn't mean the change in one variable is a result of another one.
Causation means one variable is directly affected
by another.
2. Simple Linear Regression
regression analysis that shows relationship between X and Y variables
3. X Variable
Y Variable
X= Independent and Explanatory Variable
Y= Dependent and Response Variable
4. Ordinary Least Squares
(OLS)
- What does this do?
- What does it minimize?
- What do OLS Formulas
produce?
- What point does the
OLS regression line always pass through on a
scatter plot?
Estimates coefficients of Slope and Intercept to
ensure best fit
5. Multiple Linear Regression
regression analysis that shows relationship between multiple independent X variables and a
dependent Y variable
Minimizes Residuals
OLS Formulas produce unbiased and consistent
estimates of slope and intercept
The OLS Reg. Line always passes through x(bar),
y(bar)
6. What is 5?
Random Error
What does it represent?
How is it distributed?
Represents everything not included in the model
Is it observed?
What type of Variance? Normally Distributed
Are they independent?
1/4
BUAD 310: Final Exam
Study online at https://quizlet.com/_5rrk1n
Unobservable
Constant Variance
Independent
7. How do the Population
Slope and Intercept relate
to Slope and Intercept of
Least Square Line
Utilize the Slope and Intercept of the Least
Squares Line to obtain an accurate estimate of
the Population Slope and Intercept which reduces
residuals making it a better fit
8. Residual Plot
- What does it tell us?
- What does the fan-out
pattern mean?
- What does the curved
pattern mean?
If points are randomly dispersed around the x
axis, then a linear regression model is appropriate
for the data
Fan out pattern means increased residual variance
Curved pattern means nonlinear
9. ANOVA Table
A table used to summarize the analysis of
variance computations and results. It contains
columns showing the source of variation, the sum
of squares, the degrees of freedom, the mean
square, and the F value(s).
10. What is the St. Dev. of the The Population St. Dev.
Error Term?
11. What is the Standard Er- Overall measure of model fit
ror?
What does a Standard Er- =0 means Perfect fit aka. strong correlation
ror = 0 mean?
12. Variation Around Mean of
Y
- What is SST, SSR, and
SSE
- What does a good and
perfect fit look like?
SST= Total Variation around the mean of Y
SSR= Explained Variation in Y which is difference
between conditional and unconditional mean
SSE= Unexplained Variation in Y
2/4
BUAD 310: Final Exam
Study online at https://quizlet.com/_5rrk1n
Good Fit= SSE<SST
Perfect Fit= SSE= 0 = conditional=unconditional
mean
13. How do you solve issues of Heteroscedasticity (Non-Constant) in
Residual Graphs?
Transform X and Y by taking Logs
14. What happens to
Rsquared when a new
explanatory variable is
added to a regression?
It increases because in a regression, the errors
are minimized thus maximizing rsquared. Therefore, when you increase k, the quality of the fit is
improved thus increasing rsquared.
15. Adjusted Rsquared
What does it get rid of?
What does a small/large
gap mean between
rsquared and adj.
rsquared?
Add additional predictors to increase rsquared
Gets rid of useless predictors
Small Gap: Parsimonious Model- model with no
useless predictors
Large Gap: Lean model obtained without losing
predictive power
16. F-Test
Test regression for significance
- What is it?
- What does it compare? it compares explained variation in Y (SSR) and
unexplained variation in Y (SSE)
17. Multicollinearity
- What is it?
- What does it cause?
- How to mitigate its effects?
X's are intercorrelated instead of being independent
Variance Inflation
- Results in wider cond. interval
- Makes T Stat. less reliable
Use Variance Inflation Factor (VIF)
- Tests for multicollinearity in a model
- If X is independent, its VIP=1 and R=0
3/4
BUAD 310: Final Exam
Study online at https://quizlet.com/_5rrk1n
- The larger the VIP, the more the predictor
shouldn't be in the model
18. Normal Distribution
Characteristics
Symmetrical
Unimodal
68% of Data Within 1 SD of Mean
95% of Data Within 2 SD of Mean
99.7% Data Within 3 SD of Mean
19. Robust: Definition
Strength of a model, statistic, etc.
20. Central Limit Theorem
The sampling distribution of the mean of any variable will be normal
21. Population Proportion
Fraction of pop. with certain characteristic
- What is it?
- What is used to estimate Sample Proportion
it?
22. Standard Error Of Sample Mean
Standard Error Of The
Sample Proportion
the standard deviation of the distribution of sample means
23. 1 Tailed Vs. 2 Tailed Tests 1 Tailed: Determine differences amongst groups
in a specific aspect
2 Tailed: See if there is difference in between
groups being compared
24. Type 1 and 2 Errors
Type 1: Rejecting Null when it's true
Type 2: Failure to reject Null when its false
4/4
Download