Clustered Standard Errors in Heckprobit

[Christine Braun]

I am running a heckprobit for selection and would like to multi-way cluster the standard errors on states and year. It seems like all the multi-way clustering packages are not compatible with heckprobit (I may be wrong about this). I guess the solution would be to run the heckporbit clustering by state, year, and stateXyear. Then could one calculate the variance of multi-way cluster by simply doing: V_state + V_year - V_stateXyear?

[Phil Bromiley]

You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

You might look at user-written cmp or the selection models available in GSEM - they may handle the problem. I can't say for sure, but I would be surprised if the solution to the clustered standard error problem for a probit with selection is as simple as you suggest.