Test for autocorrelation with panel data and interactions terms

[Rick Mould]

Hi everyone:

I have a new little problem I can't find any solution for in prior topics. I want to test for autocorrelation and wanted to use the Durbin-Watson test. However, since I'm using panel data, I get the error message "sample may not include multiple panels" when using dwstat. With the help of prior topics I got aware of the xtserial (ssc) command but my regression includes interaction terms so that I get the error message "factor variables and time-series operators not allowed".

Is there any other possibility to test for autocorrelation when using panel data and interaction terms?

Thank you in advance,
Rick

[Kristen Rucki]

Hi Rick,

This won't help, but wanted to let you know I've been searching through Statalist today with the exact same question. I'll let you know if I find any other options.