Question 1: Regression model selections. Using QOL as the outcome and age1, treatment, and severity as the candidate predictor variables, conduct a stepwise model selection procedure (8 pts)

1. Using SLEntry = 0.01 and SLStay = 0.05, what are variables being selected? (2 pts) 2. Describe this model selection procedure from step 1 to the final step (2 pts) 3. What is the R squared and CP value for the final model? (2 pts) 4. Check if there are any collinearity problems for the final model? (2 pts)

Forward Selection: Step 1 Forward Selection: Step 2 Variable Severity Entered: R Square - 0.2227 and C(p) - 55.3164 Variable Treatment Entered: R Square - 0.2507 and C[p) - 26 6690 Analysis of Variance Sum of Mean Analysis of Variance Source OF Squares Square F Value Pis F Sum of Mean Model 26780 26780 228 32 - 0001 Source OF Squares Square F Value Pr > F Error 797 93485 117 29580 Model 30150 15075 133 16: <:0001 corrected total comected parameter standard variable estimate error type value prat intercept entimate etfor f pro severity . treatment bounds on condition number number: forward selection: step age entered: r-square and c analysis of variance sum mean source squares square pre all variables have been entered into the model. model summary selection ii ss pr partial vars in r> F Intercept 58 28567 4 98647 15021 136 63 0001 1 Severity 0 2227 0 2227 55.3164 228 32 F Model 3 32862 10954 99.64 <.0001 error corrected total parameter standard variable estimate type ii ss f value pr s intercept age1 treatment severity bounds on condition number: all variables left in the model are significant at level>