Inference in Panel Data Models via Gibbs Sampling
In this chapter we consider the use of the recently developed Gibbs sampling method to estimate panel data models. The Gibbs sampler is a Markov chain Monte-Carlo (MCMC) method that provides an approach to simulating a given joint distribution. Although this method can be employed quite generally it has proved most useful in Bayesian inference where it has been used to simulate posterior distributions in a number of different settings (Geman and Geman , Gelfand and Smith , Tierney , and Chib and Greenberg ). Once a sample of parameter draws from the posterior distribution has been obtained it is possible to estimate a parameter of interest by taking empirical averages of the simulated values.
KeywordsPosterior Distribution Gibbs Sampling American Statistical Association Markov Chain Monte Carlo Method Panel Data Model
Unable to display preview. Download preview PDF.
- Albert, J. and S. Chib : Bayesian probit modeling of binary repeated measures data with an application to a cross—over trial, in Bayesian Biostatistics(eds. D. A. Berry and D. K. Stangl), New York: Marcel Dekker, forthcoming.Google Scholar
- Best, D. J. : Letter to the Editor, Applied Statistics, 29, 181.Google Scholar
- Chib, S. and E. Greenberg : Markov chain Monte Carlo methods in econometrics, John M. Olin School of Business, Washington University, St. Louis.Google Scholar
- Geweke, J. : Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments, Proceedings of the Fourth Valencia International Conference on Bayesian Statistics, (eds., J. M. Bernardo, J. O. Berger, A. P. Dawid, and A. F. M. Smith), New York: Oxford University Press, 169–193.Google Scholar
- Heckman, J.J. : Statistical models for discrete panel data, in Structural Analysis of Discrete Data with Econometric Applications, ed C. F. Manski and D. McFadden, pp 114–178, Cambridge: MIT Press.Google Scholar
- Roberts, G. O. and Smith, A. F. M. : Some convergence theory for Markov chain Monte Carlo, manuscript.Google Scholar
- Smith, A. F. M. and G. O. Roberts : Bayesian computation via the Gibbs sampler and related Markov chain Monte Carlo methods, Journal of the Royal Statistical Society, B, 55, 3–24.Google Scholar
- Stout, W. F. : Almost Sure Convergence, New York, Academic Press.Google Scholar
- Tierney, L. : Markov chains for exploring posterior distributions, manuscript.Google Scholar