I want to run a regression considering fixed effect, with different slope and intercept for industry and year(i.e: each industry in each year has its slope and intercept, so if there are five year and ten industry, there would be fifty different intercepts and slopes), I tried the following code, but there's only one intercept.
////
proc glm data=ipot3;
class industry year;
model y = x1*industry*year x2*industry*year x3*industry*year / solution;
run;
quit;
////
Is it possible to generate their unique intercepts?
@spasico wrote:
I want to run a regression considering fixed effect, with different slope and intercept for industry and year(i.e: each industry in each year has its slope and intercept, so if there are five year and ten industry, there would be fifty different intercepts and slopes), I tried the following code, but there's only one intercept.
////
proc glm data=ipot3;
class industry year;
model y = x1*industry*year x2*industry*year x3*industry*year / solution;
run;
quit;////
Is it possible to generate their unique intercepts?
proc glm data=ipot3;
class industry year;
model y = industry*year x1*industry*year x2*industry*year x3*industry*year / solution noint;
run;
quit;
the term Industry*Year will be the 50 intercepts.
I see now. Thanks a lot!
It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.
Ready to level-up your skills? Choose your own adventure.