Estimating Parameters of Structural Models Using Neural Networks

37 Pages Posted: 9 Dec 2019 Last revised: 30 May 2020

See all articles by Yanhao 'Max' Wei

Yanhao 'Max' Wei

University of Southern California - Marshall School of Business

Zhenling Jiang

Georgia State University - Department of Marketing

Date Written: May 21, 2020

Abstract

Machine learning tools such as neural networks see increasing applications in marketing and economics for predictive tasks, such as classifying images and forecasting choices. Instead of these predictive tasks, we explore using neural nets to estimate the parameter values for an economic model. The neural net is trained with model-generated datasets. Through training, the neural net learns a direct mapping from (the moments of) a dataset to the parameter values under which the dataset is generated. We show this Neural Net Estimator (NNE) converges to Bayesian parameter posterior when the number of training datasets is sufficiently large. We examine the performance of NNE in two Monte Carlo studies. NNE incurs substantially smaller simulation costs compared to simulated MLE and GMM, while achieving no worse estimation accuracy. NNE is also easy to implement with the wide availability of neural net training packages.

Keywords: structural estimation, computational costs, neural networks, machine learning, market entry, choice models.

Suggested Citation

Wei, Yanhao and Jiang, Zhenling, Estimating Parameters of Structural Models Using Neural Networks (May 21, 2020). Available at SSRN: https://ssrn.com/abstract=3496098 or http://dx.doi.org/10.2139/ssrn.3496098

Yanhao Wei (Contact Author)

University of Southern California - Marshall School of Business ( email )

701 Exposition Blvd
Los Angeles, CA 90089
United States

Zhenling Jiang

Georgia State University - Department of Marketing ( email )

United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
81
Abstract Views
526
rank
323,858
PlumX Metrics