J. Econometrics Vol. 228 (1) pp. 27-38
a A. Ronald Gallant
aDepartment of Economics, Penn State University,
University Park PA 16802, USA
Received 23 January 2020, Revised 1 August 2020, Accepted 9 February 2021, Available online 10 March 2021, Version of Record 11 March 2022.
Nonparametric Bayesian estimation subject to overidentified moment equations is a challenge because the support of the posterior is a manifold of lower dimension than the number of model parameters. The manifold therefore has Lebesgue measure zero thus inhibiting the use of the most commonly used Bayesian estimation method: MCMC (Markov Chain Monte Carlo). This study proposes an effective MCMC algorithm and algorithms for estimating scale and the normalizing constant. The algorithms are illustrated with two illustrative applications.
JEL Classification: C11, C14, C15, C32, C36, C58
Keyword(s): Method of moments, Bayesian inference