Taylor & Francis Group
Browse
ubes_a_1668799_sm7274.zip (2.37 MB)

Gaussian Processes and Bayesian Moment Estimation

Download (2.37 MB)
Version 3 2021-09-29, 15:53
Version 2 2019-12-09, 12:19
Version 1 2019-09-19, 15:06
dataset
posted on 2021-09-29, 15:53 authored by Jean-Pierre Florens, Anna Simoni

Given a set of moment restrictions (MRs) that overidentify a parameter θ, we investigate a semiparametric Bayesian approach for inference on θ that does not restrict the data distribution F apart from the MRs. As main contribution, we construct a degenerate Gaussian process prior that, conditionally on θ, restricts the F generated by this prior to satisfy the MRs with probability one. Our prior works even in the more involved case where the number of MRs is larger than the dimension of θ. We demonstrate that the corresponding posterior for θ is computationally convenient. Moreover, we show that there exists a link between our procedure, the generalized empirical likelihood with quadratic criterion and the limited information likelihood-based procedures. We provide a frequentist validation of our procedure by showing consistency and asymptotic normality of the posterior distribution of θ. The finite sample properties of our method are illustrated through Monte Carlo experiments and we provide an application to demand estimation in the airline market.

Funding

The authors gratefully thank the Editor, an Associate Editor, and three anonymous referees for their many constructive comments on the previous version of the article. The authors are grateful to Nicolas Chopin, Yuichi Kitamura, Frank Kleibergen, Andriy Norets and seminars and conferences participants at: Berlin, Boston College, Bristol, Carlos III, CREST, Leiden, Northwestern, SBIES 2015, ICEEE 2015, CFE - CMStatistics 2015, NASM 2012, Toulouse, for useful comments. We thank financial support from ANR-13-BSH1-0004 (IPANEMA). Anna Simoni gratefully acknowledges financial support from Labex ECODEC (ANR - 11-LABEX-0047), SFB-884 and hospitality from the University of Mannheim.

History

Usage metrics

    Journal of Business & Economic Statistics

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC