Calculate the correlation between fitted and observed using the following procedure:

The SAS System

proc reg data=hw3.mentalimp; model impair= ses life; output out=fitreg predicted=pred; run; quit;

proc corr data=fitreg; var pred impair; run; The REG Procedure Model: MODEL1 Dependent Variable: impair Number of

Observations Read 40 Number of Observations Used 40 Analysis of Variance Source DF Sum of Squares Mean Square F Value Pr > F Model 2 394.23840 197.11920 9.49 0.0005 Error 37 768.16160 20.76112 Corrected Total 39 1162.40000 Root MSE 4.55644 R-Square 0.3392 Dependent Mean 27.30000 Adj R-Sq 0.3034 Coeff Var 16.69025 Parameter Estimates Variable DF Parameter Estimate Standard Error t Value Pr > |t| Intercept 1 28.22981 2.17422 12.98 <.0001 ses 1 -0.09748 0.02908 -3.35 0.0019 life 1 0.10326 0.03250 3.18 0.0030 -------------------------------------------------------------------------------- The SAS System The REG Procedure Model: MODEL1 Dependent Variable: impair Panel of fit diagnostics for impair. Panel of scatterplots of residuals by regressors for impair. -------------------------------------------------------------------------------- The SAS System The CORR Procedure 2 Variables: pred impair Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum Label pred 40 27.30000 3.17941 1092 20.25417 35.21267 Predicted Value of impair impair 40 27.30000 5.45941 1092 17.00000 41.00000 Pearson Correlation Coefficients, N = 40 Prob > |r|