Statistics Homework Solutions 198

# Statistics Homework Solutions 198 - 198 CHAPTER 8(f...

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 198 CHAPTER 8 (f) 1683.5 (g) (1634.7, 1732.2) P 0 414 0 425 ¨ ¨ ¨ ¨ T 11 686 3 6197 P 0 000 0 007 ¨ ¨ ¨ ¨ ¨ ¨ ¨ ¨ ¡ ¡ ¨ ¨ StDev 3 4545 0 0043503 ¨ ¨ ¨ ¡ Coef 40 370 0 015747 ¨ T 0 86215 0 83993 ¨ ¨ ¨ StDev 16 754 0 20637 P 0 044 0 180 0 006 ¨ Predictor Constant x2 ¨ ¨ ¡ (c) Coef 14 444 0 17334 T 2 4572 1 4885 3 9164 ¨ Predictor Constant x1 StDev 10 424 0 12353 0 0040542 ¨ (b) Coef 25 613 0 18387 0 015878 ¨ Predictor Constant x1 x2 ¨ 15. (a) (d) The model containing x2 as the only independent variable is best. There is no evidence that the coefﬁcient of x1 differs from 0. The model y β0 β1 x2 ε is a good one. One way to see this is to compare the ﬁt of this model to the full quadratic model. The ANOVA table for the full model is ¢ SS 4.1007 3.9241 8.0248 The ANOVA table for the model y Source Regression Residual Error Total DF 1 13 14 ¡ DF 5 9 14 MS 0.82013 0.43601 β0 ε is ¢ ¡ Source Regression Residual Error Total β1 x 2 SS 2.7636 5.2612 8.0248 ¢ ¢ MS 2.7636 0.40471 F 1.881 P 0.193 F 6.8285 P 0.021 From these two tables, the F statistic for testing the plausibility of the reduced model is 0 7667. ¤ ©¥ £ £ ¨ ¡ ¨ ¤ ¨ The null distribution is F4 9 , so P 0 10 (a computer package gives P that the reduced model is plausible. ¡ 1 0 573). The large P-value indicates ¨ ¡ 5 ¥ 3 9241 3 9241 9 ¡ 5 2612 ¨ ¨ ¡ ¡ 17. ...
View Full Document

## This note was uploaded on 12/20/2011 for the course STA 3163 taught by Professor Mattgilg during the Fall '11 term at UNF.

Ask a homework question - tutors are online