I am using sfpanel() command and I need to set initial values for the variables before running the regression.
In order to set the initial values, I am using the ‘svfrontier’ argument of sfpanel().
Since I’m including user fixed effects (and there are ~2000 Users in the dataset), I need to specify around 2000 starting values. This becomes a very long expression!
Command: Sfpanel y dummy2 dummy3 i.userid, model(tfe) dist (tnormal) svfrontier( 1 2 3 … 2000 )
The numbers inside the bracket (i.e. 1 2 3…) are the starting values here (say).
I’m getting the following Error:
expression too long
r(130);
Things I've tried: 1.) I have tried this for a smaller dataset (that had 15 users) and this worked perfectly fine but for the complete dataset (~2000 users), we've a longer expression and this becomes a problem. 2.) I tried the delimit command in the do file as well:
#delimit ; sfpanel y dummy2 dummy3 i.userid, model(tfe) dist(tnormal) svfrontier(1 2 3 ... 2000); #delimit cr This didn't work for the complete dataset. (although it did work for the smaller dataset, so shouldn't be a syntax error)
Any help will be greatly appreciated! Thanks, Aishvarya
In order to set the initial values, I am using the ‘svfrontier’ argument of sfpanel().
Since I’m including user fixed effects (and there are ~2000 Users in the dataset), I need to specify around 2000 starting values. This becomes a very long expression!
Command: Sfpanel y dummy2 dummy3 i.userid, model(tfe) dist (tnormal) svfrontier( 1 2 3 … 2000 )
The numbers inside the bracket (i.e. 1 2 3…) are the starting values here (say).
I’m getting the following Error:
expression too long
r(130);
Things I've tried: 1.) I have tried this for a smaller dataset (that had 15 users) and this worked perfectly fine but for the complete dataset (~2000 users), we've a longer expression and this becomes a problem. 2.) I tried the delimit command in the do file as well:
#delimit ; sfpanel y dummy2 dummy3 i.userid, model(tfe) dist(tnormal) svfrontier(1 2 3 ... 2000); #delimit cr This didn't work for the complete dataset. (although it did work for the smaller dataset, so shouldn't be a syntax error)
Any help will be greatly appreciated! Thanks, Aishvarya
Comment