data oecd; infile "C:\Users\joann\Desktop\ECON4220\Japan.txt" dsd missover dlm="09"X; input cons gov hours inc ln_y_c r; run; proc print data =oecd; var cons gov hours inc r; run; proc reg data=oecd; model cons= hours gov r inc; run; * restricted estimation with beta_2=berta_3=0; proc reg data=oecd; model cons= hours gov r inc; restrict hours=0, gov=0; run; * test beta_2=berta_3=0; proc reg data=oecd; model cons= hours gov r inc; test hours=0, gov=0; run; * is there collinearity?; proc reg data=oecd; model cons= hours gov r inc/vif; run; proc corr data=oecd; var hours gov r inc; run; * without gov; proc reg data=oecd; model cons= hours r inc; run; * gov regressed on other explanatory; proc reg data=oecd; model gov= hours r inc; run; * to make prediction of last year I remove it from data; data oecd; point=nobs; modify oecd nobs=nobs point=point; remove; stop; run; data oecd2; input cons gov hours inc r; datalines; . 0.016254 0.005045 -0.003768 0.005812 ; run; data oecd2; set oecd oecd2; run; proc reg data=oecd2; model cons= hours gov r inc/p cli; title "prediction and prediction interval"; run; proc reg data=oecd2; model cons= hours r inc/p cli; title "prediction and prediction interval"; run; 

Does RESET suggest the 6.31 Everaert and Pozzi" develop a model to examine the predictability of consumption growth in 15 OECD countries. Their data is stored in the file ceed. The variables used are growth in real per capita private consumption (CSUMPTN), growth in real per capita goverment consumption (GOV), growth in per capita hours worked (HOURS), growth in per capita real disposable labor income (INC), and the real interest rate (R). Using only the data for Japan, answer the following questions: a. Estimate the following model and report the results CSUMPTN = B. + B, HOURS + B.GOV +PR+ByINC + Are there any coefficient estimates that are not significantly different from zero at a 5% level? b. The coefficient, could be positive or negative depending on whether hours worked and private consumption are complements or substitutes. Similarly, B, could be positive or negative depending on whether government consumption and private consumption are complements or substitutes. What have you discovered? What does a test of the hypothesis HB, = 0, B = 0 reveal? c Re-estimate the equation with GOV omitted and, for the coefficients of the remaining variables, comment on any changes in the estimates and their significance. d. Estimate the equation GOV = 4+ HOURS + a,R + .INC + v and use these estimates to reconcile the estimates in part (a) with those in part (e)