Working Paper: CEPR ID: DP18560
Authors: Hyungsik Roger Moon; Frank Schorfheide; Boyuan Zhang
Abstract: We incorporate a version of a spike and slab prior, comprising a pointmass at zero ("spike") and a Normal distribution around zero ("slab") into a dynamic panel data framework to model coefficient heterogeneity. In addition to homogeneity and full heterogeneity, our specification can also capture sparse heterogeneity, that is, there is a core group of units that share common parameters and a set of deviators with idiosyncratic parameters. We fit a model with unobserved components to income data from the Panel Study of Income Dynamics. We find evidence for sparse heterogeneity for balanced panels composed of individuals with long employment histories.
Keywords: Bayesian analysis; Forecasting; Income dynamics; Panel data models; Sparsity; Spike-and-slab priors
JEL Codes: C11; C23; C53; E20
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| spike-and-slab prior (C11) | modeling of coefficient heterogeneity (C51) |
| modeling of coefficient heterogeneity (C51) | capture sparse heterogeneity in income dynamics (J69) |
| sparse heterogeneity in balanced panels (C23) | similar income profiles (D31) |
| unbalanced panels (C23) | full heterogeneity (B50) |
| return to experience (Y60) | sparse heterogeneity in income dynamics (D31) |
| autocorrelation of persistent income components (C22) | sparse heterogeneity in income dynamics (D31) |