Hi,
I have to run approx 2500 separate regressions. I'm having trouble getting ONLY the output I want in one nice output file. An excerpt of my code is below.
proc reg data=chris outest=est1 tableout;
model1: model rpfund1=rpfifty/ RSQUARE noprint;
run;
proc reg data=chris outest=est2 tableout;
model2: model rpfund2=rpfifty/RSQUARE noprint;
run;
proc print data=est1;
run;
proc print data=est2;
run;
I only want the R squared value, the parameter estimates (for both the independent variable and the intercept) the t-statistics and the MSE and the number of observations used. I get all of this with the above code except the number of observations used in the regression. Anyone know how to report this? With the above code confidence intervals are also reported which I don't want.
The main problem is the outputs are in 2500 different tables.
I tried to use proc append to get all of the output in one table and then export it into Excel but was unsuccessful.
Can any one help please?
Thanks very much,
Chris
I have to run approx 2500 separate regressions. I'm having trouble getting ONLY the output I want in one nice output file. An excerpt of my code is below.
proc reg data=chris outest=est1 tableout;
model1: model rpfund1=rpfifty/ RSQUARE noprint;
run;
proc reg data=chris outest=est2 tableout;
model2: model rpfund2=rpfifty/RSQUARE noprint;
run;
proc print data=est1;
run;
proc print data=est2;
run;
I only want the R squared value, the parameter estimates (for both the independent variable and the intercept) the t-statistics and the MSE and the number of observations used. I get all of this with the above code except the number of observations used in the regression. Anyone know how to report this? With the above code confidence intervals are also reported which I don't want.
The main problem is the outputs are in 2500 different tables.
I tried to use proc append to get all of the output in one table and then export it into Excel but was unsuccessful.
Can any one help please?
Thanks very much,
Chris