A Pre-Trained Estimator for Consumer Search Model

26 Pages Posted: 7 Jun 2024

See all articles by Yanhao 'Max' Wei

Yanhao 'Max' Wei

University of Southern California - Marshall School of Business

Zhenling Jiang

University of Pennsylvania - The Wharton School

Date Written: June 08, 2024

Abstract

We explore pre-training estimators for structural econometric models. The estimator is "pre-trained" in the sense that most computations and effort occur once during its construction. Subsequent applications of it to estimate different datasets require little computation costs or effort. The estimation relies on a neural net to recognize structural model's parameter from data patterns. This paper focuses on a sequential search model that is known to be hard to estimate. We evaluate our pre-trained estimator on 11 real datasets. The estimation takes seconds to run and shows high accuracy. We provide it at pnnehome.github.io. More generally, pre-trained estimators make structural models more accessible and easier to apply. They can also facilitate privacy-preserving estimation because they need only aggregate data patterns.

Suggested Citation

Wei, Yanhao and Jiang, Zhenling, A Pre-Trained Estimator for Consumer Search Model (June 08, 2024). Available at SSRN: https://ssrn.com/abstract=4856490 or http://dx.doi.org/10.2139/ssrn.4856490

Yanhao Wei (Contact Author)

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

701 Exposition Blvd
Los Angeles, CA California 90089
United States

Zhenling Jiang

University of Pennsylvania - The Wharton School ( email )

3641 Locust Walk
Philadelphia, PA 19104-6365
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
26
Abstract Views
136
PlumX Metrics