Regime switching panel data models with interative fixed effects

In this paper, we introduce a regime switching panel data model with interactive fixed effects. We propose a maximum likelihood estimation method and develop an expectation and conditional maximization algorithm to estimate the unknown parameters. Simulation results show that the algorithm works well in finite samples. The biases of the maximum likelihood estimators are negligible and the root mean squared errors of the maximum likelihood estimators decrease with the increase of either cross-sectional units N or time periods T.