Scandinavian Working Papers in Economics

SSE/EFI Working Paper Series in Economics and Finance,
Stockholm School of Economics

No 353: Panel Regression with Unobserved Classes

Mickael Salabasis () and Mattias Villani ()
Additional contact information
Mickael Salabasis: UC AB, Postal: Analyssektionen, SE-117 88 Stockholm, Sweden
Mattias Villani: Dept. of Statistics, Stockholm University, Postal: Stockholm University, S-106 91 Stockholm, Sweden

Abstract: We propose a panel regression model with a predetermined and fixed number of classes, where each class is defined by its parameters, but any reference as to which group any observation belongs to is absent. The classes or groups are rationalized by a willingness to attribute some of the observed heterogeneity on a higher level than the individual. The estimation procedures have a distinct Bayesian flavor, relying on the Gibbs sampler for parameter estimation, a method proven effective in situations with missing or latent variables.

Keywords: Panel data; Bayesian statistics

JEL-codes: C11; C33

19 pages, January 24, 2000

Full text files PostScript file Full text PostScript file Full text
hastef0353.pdf PDF-file Full text PDF-file Full text

Download statistics

Questions (including download problems) about the papers in this series should be directed to Helena Lundin ()
Report other problems with accessing this service to Sune Karlsson ().

This page generated on 2018-03-27 10:24:47.