Hello,

I am using panel data, where my dataset includes 16 different target firm characteristics and the dependent variable is a dummy that equals 1 in case an investment fund has taken a 1% stake in the target firm. I have 15 years of data on approximately 8000 firms. My version of Stata is 14.0.

When I run a regression without lags everything works fine (I need to use clustered standard errors to account for heteroskedasticity):

Code:
 xtprobit DOwn1 ESG FSize Leverage ROA , re vce(cluster ID)
However, when I am trying to include the lag of 'ESG' (as below), the error message: "calculation of robust standard errors failed" (r198) comes up. This also happens when I include lags of other independent variables.

Code:
 xtprobit DOwn1 L.ESG FSize Leverage ROA , re vce(cluster ID)
I cannot seem to figure out how to solve the problem; except by manually creating a new variable that is the lag of 'ESG', and using that instead of L.ESG. This method does not spark an error message; however I am hesitant about whether I am overlooking some error that leads to my output being wrong.

Thus my questions: is there a fix that allows me to calculate the robust standard errors while including lags or is my manual 'fix' a good way around the problem?

Thanks in advance for any thoughts!