Dynamic GMM with dummy variables
Posted: Thu Jul 02, 2015 4:33 am
Hello,
I'm attempting to run a Dynamic Panel Regression with cross sectional differences. I want to include a dummy variable, which would obviously be differenced. I was under the impression that I could include the variable in the instruments list as @lev(dummy), but each time I run the regression I'm given a near singular matrix error. Is there any way around this? I'm not particularly familiar with GMM but I'm trying to replicate a study in a paper which ran a GMM regression with a lagged dependent variable, so I assume it would be using differences.
Any help you're able to offer would be greatly appreciated.
Regards,
Daniel.
I'm attempting to run a Dynamic Panel Regression with cross sectional differences. I want to include a dummy variable, which would obviously be differenced. I was under the impression that I could include the variable in the instruments list as @lev(dummy), but each time I run the regression I'm given a near singular matrix error. Is there any way around this? I'm not particularly familiar with GMM but I'm trying to replicate a study in a paper which ran a GMM regression with a lagged dependent variable, so I assume it would be using differences.
Any help you're able to offer would be greatly appreciated.
Regards,
Daniel.