Neural nets for indirect inference
Creel, Michael (Universitat Autònoma de Barcelona. Departament d'Economia i d'Història Econòmica)

Imprint: Universitat Autònoma de Barcelona. Unitat de Fonaments de l'Anàlisi Econòmica 2016
Description: 15 p.
Abstract: This paper shows how neural networks may be used to approximate the limited information posterior mean of a simulable model. Because the model is simulable, training and testing samples may be generated with sizes large enough to train well a net that is large enough, in terms of number of hidden layers and neurons, to learn the limited information posterior mean with good accuracy. The output of the net can be used as an estimator of the parameter, or, following Jiang et al. (2015), as an input to subsequent classical or Bayesian indirect inference estimation. Targeting the limited information posterior mean using neural nets is simpler, faster, and more successful than is targeting the full information posterior mean. Code to replicate the examples and to use the methods for other models is available at https://github. com/mcreel/NeuralNetsForIndirectInference. jl. This code uses the Mocha. jl package for the Julia language, which allows for easy access to GPU computing, which greatly accelerates training the net.
Rights: L'accés als continguts d'aquest document queda condicionat a l'acceptació de les condicions d'ús establertes per la següent llicència Creative Commons: Creative Commons
Language: Anglès.
Series: Working papers
Series: Working papers ; 960.16
Document: workingPaper
Subject: Ciències socials ; Mètodes estadístics ; Machine learning ; Indirect inference ; Neural networks ; Approximate bayesian computing

Adreça alternativa: https://hdl.handle.net/2072/260477


15 p, 509.9 KB

The record appears in these collections:
Research literature > Working papers

 Record created 2016-04-06, last modified 2019-02-02



   Favorit i Compartir