# 2SLS probit vs LPM

I am using 2SLS to estimate the effect of education on the probability that one works. In the first stage I regress education on my instrument and the other exogenous control variables. The same exogenous control variables are then included in the second stage.

The LPM version is obtained in Stata by the following command:

ivregress 2sls emp i.country i.cohort (education=instrument)


However, I cannot decide whether to use probit instead. From the literature I have mostly found support for probit. I hence wonder when LPM is consistent and/or preferred to probit?

• You are using ivprobit for the probit 2sls? The procedure you are describing in the beginning of your question is not consistent if this is what you are doing. – Andy Mar 27 '15 at 12:44
• Yes when using probit I run: ivprobit emp i.country i.cohort (education=instrument) Why would this not be consistent? – M Johnson Mar 27 '15 at 12:52
• Using ivprobit is fine. – Andy Mar 27 '15 at 12:53
• What does LPM stand for? – Richard Hardy Mar 27 '15 at 12:57
• So if I understand it correctly, it's fine to use both LPM (linear probability model) and probit in the case I have described? – M Johnson Mar 27 '15 at 13:00

For instance, if you are interested in prediction then LPM will be no good as predicted probabilities are not restricted to lie between zero and one. If you have clusters in your standard errors (in your case people in the same regions are likely to be subjected to similar shocks to their employment status), the standard errors are more easily adjusted in LPM. IV probit on the other hand is much more expensive in terms of computation and you also need to calculate the marginal effects in order to get interpretable coefficients - in Stata you can do this with the margins command.