Temporal-Difference estimation of dynamic discrete choice models

TL;DR

This paper introduces fast, flexible algorithms based on Temporal-Difference learning for estimating parameters in dynamic discrete choice models, especially effective in high-dimensional or continuous state spaces.

Contribution

It develops two novel TD-based algorithms that avoid specifying transition densities and can be combined with existing estimation methods for improved efficiency.

Findings

Algorithms are computationally efficient and scalable.

Monte Carlo simulations validate the effectiveness of the methods.

Approaches work well with high-dimensional and continuous state spaces.

Abstract

We study the use of Temporal-Difference learning for estimating the structural parameters in dynamic discrete choice models. Our algorithms are based on the conditional choice probability approach but use functional approximations to estimate various terms in the pseudo-likelihood function. We suggest two approaches: The first - linear semi-gradient - provides approximations to the recursive terms using basis functions. The second - Approximate Value Iteration - builds a sequence of approximations to the recursive terms by solving non-parametric estimation problems. Our approaches are fast and naturally allow for continuous and/or high-dimensional state spaces. Furthermore, they do not require specification of transition densities. In dynamic games, they avoid integrating over other players' actions, further heightening the computational advantage. Our proposals can be paired with popular existing methods such as pseudo-maximum-likelihood, and we propose locally robust corrections for the latter to achieve parametric rates of convergence. Monte Carlo simulations confirm the properties of our algorithms in practice.