Mundlak Approach/Interaction Term

[Guest]

Hi,

I have used the Mundlak approach (correlated random effects) to regress my dependent variable income on race (my variable of interest) and several control factors. Also, I have included an interaction term of the categorical variable: religion and the continuous variable: number of dependent children.

For my OLS and random effects regressions, I used the command: margin religion ,dydx(dependentchildren) to interpret the interaction term. However, I am unsure whether this is the correct approach for the Mundlak method as there are also meaned and demeaned variables of dependent children.

I have attached my do-file below.

Would this be the correct method and if not how can this be done?

Many thanks

[Phil Bromiley]

You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. Also, don't post files - many of us will not open files from strangers.

There are several user-written Mundlak-related estimators. Schrunk has published pieces on them in Stata Journal and the software is available. Look at Schunck, R. 2013. Within and between estimates in random-effects models: Advantages and drawbacks of correlated random effects and hybrid models. Stata Journal, 13(1): 65-76.
Schunck, R., & Perales, F. 2017. Within- and between-cluster effects in generalized linear mixed models: A discussion of approaches and the xthybrid command. Stata Journal, 17(1): 89-115