Hello,
I am interested in running a random-coefficients model via xtrc and then saving the group-specific coefficients and standard errors as new variables in the existing dataset. For example, consider a dataset consisting of variables GROUPID, Y, and X.
After setting GROUPID as the panel variable, I would run xtrc Y X, beta, which will allow me to view the group-specific coefficients and standard errors for X.
I am looking for the best way to save those group-specific coefficients as one new variable (e.g., XHAT) that will differ by group.
If I run gen XHAT = _b[X] after the regression, I only get the overall coefficient, not the group-specific ones. I see that I can save the matrices of group coefficients and variances that are saved in e(beta_ps) and e(V_ps), but I am wondering if there is a more straightforward way to save these as a new variable in the dataset instead of extracting specific cells from those matrices and ensuring that the extracted data corresponds to the proper group.
Any advice would be greatly appreciated.
I am interested in running a random-coefficients model via xtrc and then saving the group-specific coefficients and standard errors as new variables in the existing dataset. For example, consider a dataset consisting of variables GROUPID, Y, and X.
After setting GROUPID as the panel variable, I would run xtrc Y X, beta, which will allow me to view the group-specific coefficients and standard errors for X.
I am looking for the best way to save those group-specific coefficients as one new variable (e.g., XHAT) that will differ by group.
If I run gen XHAT = _b[X] after the regression, I only get the overall coefficient, not the group-specific ones. I see that I can save the matrices of group coefficients and variances that are saved in e(beta_ps) and e(V_ps), but I am wondering if there is a more straightforward way to save these as a new variable in the dataset instead of extracting specific cells from those matrices and ensuring that the extracted data corresponds to the proper group.
Any advice would be greatly appreciated.
