Latent Group Structures with Heterogeneous Distributions

This paper aims to identify the latent grouped heterogeneity of distributional effects. We consider two panel quantile regression models with additive fixed effects, where quantile slope coefficients differ across groups. The first model allows grouped and time fixed effects, and we propose a composite-quantile approach to jointly estimate group memberships, slope coefficients, and fixed effects. We show that using multiple quantiles improves clustering accuracy if memberships are quantile-invariant. Our second model extends the first specification by allowing individual and time fixed effects, and we propose an innovative double-clustering estimation approach. We apply the methods to examine the relationship between managerial incentives and risk-taking behavior