This paper considers a semiparametric panel data model with heterogeneous coefficients and individual-specific trending functions, where the random errors are assumed to be serially correlated and cross-sectionally dependent. We propose mean group estimators for the coefficients and trending...

This paper develops a method for testing for the presence of a single structural break in panel data models with unobserved heterogeneity represented by a factor error structure. The common factor approach is an appealing way to capture the effect of unobserved variables, such as skills and...