I am now running regression using proc reg function. My data set is a panel dataset. It contains 16 years data and for each year, it has thousands of observations. What I want to do is to run regression by each year first and after I get all the regression results out (16 groups of coefficients), I need to get the average of all the coefficients and R-squares. Now by the following statement, I am able to get the average of all the coefficients, but I am not able to get the average of R-squares. Is it possible to get averaged R-squares? Could anyone help me with it? Thanks very much.
ods listing close;
ods output parameterestimates=pe;
proc reg data=dset;
model depvar = indvars; run;
proc means data=pe mean std t probt;
var estimate; class variable;
Message was edited by: GingerLee
You should create a second dataset containing R-squares.
ods output parameterestimates=pe fitStatistics=fs ;
And then process it in the same way as you did with the coefficients.