Practical notes on panel data models with interactive effects

This note is intended for researchers who want to use the interactive effects model for empirical modeling. We consider how to estimate interactive effects models when some of the factors and factor loading are observable. Observable factors are common regressors which do not vary across individuals such as macroeconomic variables, but their regression coefficients are individual-dependent. Observable factor loadings correspond to time-invariant regressors such that race, gender and education, but their regression coefficients are time dependent. This note elaborates the estimation procedures in Bai (2009) in the presence of such regressors.