This preview shows pages 1–4. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 122.a)b)1218.The Minitab result is shown below.Regression Analysis: W versus GF, GA, ... The regression equation isW = 512 + 0.164 GF  0.183 GA  0.054 ADV + 0.09 PPGF  0.14 PCTG  0.163 PEN 0.128 BMI + 13.1 AVG + 0.292 SHT  1.60 PPGA  5.54 PKPCT + 0.106 SHGF+ 0.612 SHGA + 0.005 FGPredictor Coef SE Coef T PConstant 512.2 185.9 2.75 0.015GF 0.16374 0.03673 4.46 0.000GA 0.18329 0.04787 3.83 0.002ADV 0.0540 0.2183 0.25 0.808PPGF 0.089 1.126 0.08 0.938PCTG 0.142 3.810 0.04 0.971PEN 0.1632 0.3029 0.54 0.598BMI 0.1282 0.2838 0.45 0.658AVG 13.09 24.84 0.53 0.606SHT 0.2924 0.1334 2.19 0.045PPGA 1.6018 0.6407 2.50 0.025PKPCT 5.542 2.181 2.54 0.023SHGF 0.1057 0.1975 0.54 0.600SHGA 0.6124 0.2615 2.34 0.033FG 0.0047 0.1943 0.02 0.981S = 2.65443 RSq = 92.9% RSq(adj) = 86.3%Analysis of VarianceSource DF SS MS F PRegression 14 1390.310 99.308 14.09 0.000Residual Error 15 105.690 7.046Total 29 1496.000whereThe standard errors of the coefficients are listed under the SE Coef column above.1238. a) for all jfor at least one jFrom the Minitab output in Exercise 128 Reject Hand conclude that the regression model is significant at = 0.05b) GF : Reject HGA :Reject HADV : Fail to reject HPPGF : Fail to reject HPCTG :Fail to reject HPEN : Fail to reject HBMI :Fail to reject HAVG : Fail to reject HSHT : Reject HPPGA :Reject HPKPCT : Reject HSHGF :Fail to reject HSHGA : Reject HFG : Fail to reject HIt does not seem that all regressors are important. Only the regressors "GF" (β1), “GA” (β2), "SHT" (β9), “PPGA” (β10 ), "PKPCT" (β11), and “SHGA” (β13 ) are significant at = 0.05c) The Minitab result is shown below.Regression Analysis: W versus GF, PPGF The regression equation isW =  8.82 + 0.218 GF  0.016 PPGFPredictor Coef SE Coef T PConstant 8.818 9.230 0.96 0.348GF 0.21779 0.05467 3.98 0.000PPGF 0.0162 0.1134 0.14 0.888S = 5.11355 RSq = 52.8% RSq(adj) = 49.3%Analysis of VarianceSource DF SS MS F PRegression 2 789.99 395.00 15.11 0.000Residual Error 27 706.01 26.15Total 29 1496.00Because PPGF had a tstatistic near zero in part (b) there is a concern that it is not an important predictor. We will evaluate its role in the smaller model with GF.Because f> f0.05,2,27, we reject the null hypothesis that the coefficient of GF and PPGF are both zero.Reject HFail to reject HBased on the ttest, power play goals for (PPGF) is not a logical choice to add to the model that already contains GF. 1256.a) From the Minitab output in Exercise 1218 the estimate, standard error, t statistic and Pvalue for the coefficient of GF are:Predictor Coef SE Coef T PGF 0.16374 0.03673 4.46 0.000The 95% CI on the regression coefficient β1of GF isb) The Minitab result is shown below....
View
Full
Document
This note was uploaded on 11/18/2011 for the course ECE 083 taught by Professor Best during the Spring '08 term at Lehigh University .
 Spring '08
 BEST

Click to edit the document details