appiled linear regression

 appiled thready regression Composition

College student

Solutions Manual

to accompany

Applied Thready

Regression Models

Fourth Edition

Michael They would. Kutner

Emory University

Christopher J. Nachtsheim

University of Minnesota

John Neter

School of Georgia

2004

McGraw-Hill/Irwin

Chicago, ARIANNE

Boston, MOTHER

PREFACE

This kind of Student Alternatives Manual provides intermediate and final statistical results for any starred (*) end-of-chapter Complications with computational elements contained in Utilized Linear Regression M odels, 4th copy. No solutions are given pertaining to Exercises, Projects, or Circumstance Studies.

In presenting calculational results all of us frequently demonstrate, for convenience in examining, more numbers than will be significant pertaining to the original data. Students and other users may obtain slightly different answers than those provided here, because of different rolling procedures. Each time a problem takes a percentile (e. g. in the t or F distributions) not included inside the Appendix N Tables, users may either interpolate in the table or perhaps employ an available laptop program intended for finding the needed benefit. Again, a bit different beliefs may be obtained than the ones shown here.

The data sets for all Complications, Exercises, Projects and Case Studies are contained in the compact disk supplied with the text to facilitate info entry. It is expected the fact that student will use a computer and have access to computer output for all those but the most basic data sets, where use of a basic calculator would be sufficient. For most students, hands-on knowledge in getting the computations by simply computer will be an important area of the educational experience in the course.

Although we have examined the alternatives very carefully, it will be easy that several errors continue to be present. We might be the majority of grateful to acquire any problems called to the attention. Dejadez can be reported via the website for the book: http://www.mhhe.com/KutnerALRM4e. We admit with thanks the assistance of Lexin Li and Yingwen Dong in the checking of this manual. We all, of course , are in charge of for any mistakes or absences that remain. Michael L. Kutner

Christopher J. Nachtsheim

John Neter

i

2

Contents

1 LINEAR REGRESSION WITH 1 PREDICTOR VARIABLE

1-1

two INFERENCES IN REGRESSION AND CORRELATION EVALUATION

2-1

several DIAGNOSTICS AND REMEDIAL PROCEDURES

3-1

4 SIMULTANEOUS INFERENCES AND OTHER MATTERS IN REGRESSION ANALYSIS 4-1

5 MATRIX APPROACH TO STRAIGHTFORWARD LINEAR REGRESSION ANALYSIS

5-1

6 MULTIPLE REGRESSION – I

6-1

7 MULTIPLE REGRESSION – II

7-1

8 DESIGNS FOR QUANTITATIVE AND QUALITATIVE PREDICTORS 8-1

9 BUILDING THE REGRESSION MODEL I actually: MODEL SELECTION AND

APPROVAL

9-1

10 BUILDING THE REGRESSION MODEL II: ANALYSIS

10-1

14 BUILDING THE REGRESSION UNIT III: HELPFUL MEASURES

11-1

12 AUTOCORRELATION IN TIME SERIES DATA

12-1

13 SUMMARY OF NONLINEAR REGRESSION AND NERVE ORGANS NETWORKS

13-1

14 LOGISTIC REGRESSION, POISSON REGRESSION, AND GENERALIZED LINEAR MODELS 14-1

iii

4

Chapter 1

LINEAR REGRESSION WITH ONE PARTICULAR

PREDICTOR VARIABLE

1 . twenty. a.

m.

1 . twenty one. a.

n.

c.

g.

1 . twenty-four. a.

ˆ

Y sama dengan −0. 5802 + 12-15. 0352X

ˆ

Yh = 74. 5958

ˆ

Con = twelve. 20 & 4. 00X

ˆ

Yh = 13. 2

some. 0

¯ ¯

(X, Y ) = (1, 14. 2)

i:

one particular

ei: -9. 4903

a couple of

...

0. 4392...

44

1 . 4392

forty-five

2 . 4039

e2 sama dengan 3416. 377

i

Minutes Q =

b.

1 ) 25. a.

b.

1 ) 27. a.

b.

e2

i

M SE = 79. 45063,

Meters SE = 8. 913508, minutes

e1 = 1 ) 8000

e2 = 18. 6000, Meters SE = 2 . 2150, σ 2

i

ˆ

Y sama dengan 156. thirty five − 1 . 19X

ˆ

(1) b1 = −1. 19, (2) Yh sama dengan 84. 95, (3) e8 = some. 4433,

(4) M SONY ERICSSON = 66. 8

1-1

1-2

Section 2

INFERENCES IN REGRESSION

AND RELATIONSHIP ANALYSIS

2 . 5. a.

t(. 95; 43) sama dengan 1 . 6811, 15. 0352 ± 1 . 6811(. 4831), 14. 2231 ≤ β1 ≤ 15. 8473

m.

H0: β1 = zero, Ha: β1 = 0. t∗ sama dengan (15. 0352 − 0)/. 4831 sama dengan 31. 122. If |t∗ | ≤ 1 . 681 conclude H0, otherwise Anordna. Conclude St?lla till med ett. P -value= 0+

c.

Yes

g.

H0: β1 ≤ 16, Ha: β1 > 14. t∗ = (15. 0352 − 14)/. 4831 = 2 . 1428. If t∗ ≤ 1 . 681 conclude H0, normally Ha. Determine Ha. S -value=. 0189

2 . 6th. a.

t(. 975; 8)...

Popular