Fama-MacBeth regression with Newey-West correction, but panel date is not regularly spaced

[David Tsu]

My panel data is not always consecutive: there are some gaps for time series of a few IDs. I tried both of code below to estimate with time-series autocorrelation corrected for estimating standard errors:

asreg x1 x2 x3, fmb newey(3)

xtfmb y x1 x2 x3, lag(3)

However, both report errors due to the data structure. Is there any better command I can use for the purpose?

[Phil Bromiley]

You did not get a quick answer. You'll increase the chances of a useful answer by following the FAQ on asking questions - Stata code in code delimiters, readable Stata output, and sample data using dataex. Knowing the precise statement and precise error messages is essential to diagnosing what you have. You are using user-written estimators. Help on an estimator depends on someone active on the Forum using it.

I don't use these so take my suggestions as just that suggestions. If the gaps are for a few of many ids, then you might want to fill them in. Alternatively, you might want to drop the ids. Or, if you want to ignore the gaps, you could create your own "time" id. Serial correlation is of course not estimable if your time doesn't match what the estimator expects. Finally, you can always program the estimation yourself and choose how to handle gaps.