Exam 4

advertisement

Exam 4 Fall02

Use output 1 to answer the following 10 questions

1.

How many hormone types are compared?

(a) 2

(b) 3

(c) 4

(d) 5

(e) 6

2.

Which of the following is the point estimate for

, the true overall average growth?

(a) 11.50

(b) 12.75

(c) 13.90

(d) 15.30

(e) 17.50

3.

Which of the following null hypothesis would you use to investigate if the true average growth depend on the choice of hormone? (I am asking if there are any differences between them)

(a) H

0

:

1

 

2

(b)

(c)

(d)

H

0

H

0

H

0

:

:

:

:

1

2

1

 

1

2

2

1

4

1

5

(

2

 

1

3

 

3

 

4

3

 

3

 

4

4

5

5

)

5

(e) H

0

4.

Which of the following null hypothesis would you use to investigate if the true average growth applying hormone 1 different than the others?

(a) H

0

:

1

 

2

(b)

(c)

(d)

H

0

H

0

H

0

:

:

:

:

1

2

1

 

1

2

2

1

4

1

5

(

2

1

3

 

3

 

4

3

 

3

 

4

4

5

5

)

5

(e) H

0

5.

Which of the following is the point estimate for

2

-

1

(the difference between the true average growth applying hormone 2 and hormone 1)

(a) –4.75

(b) –5

(c) 1.25

(d) 4.75

(e) 5

6.

If the true standard deviation of growth is assumed to be the same applying hormone types 1 and 2, the 95% confidence interval on the difference in true average growth applying hormone 1 and hormone

2 is (-11.7568 , 1.7568). Which of the following would be the interpretation?

(a) The true average growth applying hormone 1 and 2 are different

(b) The true average growth does depend on the choice of the hormone

(c) The true average growth applying hormone 1 and 2 are not different

(d) The true average growth does not depend on the choice of the hormone

Exam 4 Fall02

7.

If the true standard deviation were assumed to be the same applying hormone types 1 and 2, which of the following test statistics would you use to investigate if the true average growth applying hormone 1 and 2 are different?

(a) –3.49

(b) -1.811

(c) 1.811

(d) 1.868

(e) 3.49

8.

Is the assumption of equal true variances applying hormone 1 and 2 are correct using the significance level of 0.05? (p-value is based on the given test statistics for each choice)

(a) Yes, the test statistics is 1.167 and the corresponding P-value is 0.7288

(b) No, the test statistics is 1.167 and the corresponding P-value is 0.7288

(c) Yes, the test statistics is 1.361 and the corresponding P-value is 0.5878

(d) No, the test statistics is 1.361 and the corresponding P-value is 0.5878

(e) Yes, the test statistics is 1.66 and the corresponding P-value is 0.798

9.

Is the assumption of equal true variances of growth using all hormone types correct using the significance level of 0.05?

(a) Yes, the test statistics is 1.167 and the corresponding P-value is 0.7288

(b) No, the test statistics is 1.167 and the corresponding P-value is 0.7288

(c) Yes, the test statistics is 1.361 and the corresponding P-value is 0.5878

(d) No, the test statistics is 1.361 and the corresponding P-value is 0.5878

(e) Yes, the test statistics is 1.66 and the corresponding P-value is 0.798

10.

What is the point estimate for the constant standard deviation when you investigate if the true average growth depend on the choice of hormone?

(a) 3.795

(b) 14.400

(c) 50.100

(d) 207.360

(e) 215.500

11.

In testing H

0

: p

1

 p

2

, the null hypothesis must be rejected in one of the following outcomes. For what result must you reject H ? (Use

=0.01 and assume you have a large sample. Possible table

0 values you may need are z

0.02

=2.05, z

0.01

=2.33, z

0.005

=2.575).

(a) z=2.53

(b) z=2.31

(c) I need to use t distribution here but critical values are not given

(d) z=-2.31

(e) z=-2.53

Use output 2 to answer the following questions

12.

Regress time on strength. Which of the following is the least squares regression line?

(a) Time = -8.78 + 18.8735

Strength

(b) Strength = -8.78 + 18.8735

Time

(c) Time = 18.8735 - 8.78

Strength

(d) Strength = 18.8735 - 8.78

Time

(e) I do not have enough information to compute it.

Exam 4 Fall02

13.

What is the predicted time when the strength was 12.5?

(a) -90.88

(b) 102.05

(c) 227.14

(d) 12.5 is not in the range of strength to compute it

(e) 12.5 is not in the range of time to compute it

14.

What is the average change in time when the strength is increased by 0.1 unit?

(a) 0.878

(b) 1.88735

(c) 8.78

(d) 18.8735

(e) I cannot compute it since 0.1 is not in the range of strength

15.

Which of the following is the possible relation between time and strength in general?

(a) They are positively related

(b) They are negatively related

(c) They are not related

16.

I am not sure if time is really related with strength in general. Which of the following null hypothesis would you use to test this?

(a) H

0

:

0

 

8 .

78

(b)

(c)

H

H

0

0

:

:

1

0

18 .

8735

0

(d)

(e)

H

H

0

0

:

:

1

0

0

8 .

78 and H

0

:

1

18 .

8735

17.

Which of the following is the Pearson’s correlation coefficient between time and strength?

(a) -0.98995

(b) -0.97800

(c) 0.97800

(d) 0.98000

(e) 0.98995

18.

Which of the following is the point estimate for the constant standard deviation for residuals?

(a) 16.62

(b) 49.86

(c) 276

(d) 352.37

(e) 124163

19.

Which of the following is the corresponding residual when the strength is 19?

(a) –19.1835

(b) –17.1835

(c) I do not have enough information to compute it.

(d) 17.1835

(e) 19.1835

20.

If I claim that

1

19 , do I have enough evidence based on the data using

=0.05 to prove that I am right? (note that the critical value for this test is 2.262)

(a) Yes

(b) No

Exam 4 Fall02

21.

I decided to take the square root of time (called it sqrttime) and regress sqrttime on strength. I called this as Model 2 where as the first one was Model 1. The R 2 computed is 94% with model 2. Which of those models would you use?

(a) Model 1

(b) Model 2

(c)

I do not know first model’s R 2 to make a comparison.

OUTPUT 1

The data on plant growth after the application of different types of growth hormone is recorded.

Hormone 1

Hormone 2

Hormone 3

Hormone 4

Hormone 5

13

21

18

7

6

17

14

15

11

11

7

20

20

18

15

14

17

17

10

8

We are trying to investigate if choice of hormone is important for the plant growth. We modeled the data using the single factor ANOVA, X ij

  i

  ij

where i=1,...,5 and j=1,...,4.

X : plant growth ij

 i

: true average plant growth for i th hormone type.

 ij

 i

5 

1

 i

/ 5 : the true overall average plant growth

: errors which are normally distributed with mean, 0 and the constant variance,

2

.

The following output is obtained using the statistical software MINITAB.

Analysis of Variance

Source DF SS MS F P

Factor 4 200.3 50.1 3.49 0.033

Error 15 215.5 14.4

Total 19 415.8

Level N Mean StDev

Hormone1 4 12.750 4.193

Hormone2 4 17.750 3.594

Hormone3 4 17.500 2.082

Hormone4 4 11.500 4.655

Hormone5 4 10.000 3.916

Bartlett's Test (normal distribution)

Test Statistic: 1.660

P-Value : 0.798

Levene's Test (any continuous distribution)

Test Statistic: 0.283

P-Value : 0.884

Exam 4 Fall02

Normal Probability Plot for growth

ML Estimates - 95% CI

99

95

90

80

70

60

50

40

30

20

10

5

ML Estimates

Mean 13.9

StDev 4.55961

Goodness of Fit

AD* 0.874

1

0 10 20

Data

OUTPUT 2

An experiment to measure the macroscopic magnetic relaxation time in crystals as a function of the strength of the external biasing magnetic field yielded the following data.

Strength 11

Time 187

12.5 15.2 17.2 19

225 305 318 367

20.8 22

365 400

24.2 25.3 27

435 450 506

29

558

Simple linear regression, Y

 

0

 

1 x

 e is used modeling this data. The parameters

0

and

1

are true values of intercept and slope. Regressing relaxation time on strength, we obtained the following output using MINITAB software.

Predictor Coef SE Coef T P

Constant -8.78 18.75 -0.47 0.651 strength 18.8735 0.8902 21.20 0.000

S = 16.62 R-Sq = 98.0% R-Sq(adj) = 97.8%

Analysis of Variance

Source DF SS MS F P

Regression 1 124163 124163 449.47 0.000

Residual Error 9 2486 276

Total 10 126650

Answer Key:

1.D 2.C 3.C 4.E 5.E 6.C 7.B 8.C 9.E 10.A 11.A 12.A

13.C 14.B 15.A 16.D 17.E 18.A 19.D 20.A 21.A

Download