Adaptive Grids for the Estimation of Dynamic Models

62 Pages Posted: 18 Sep 2020 Last revised: 21 Sep 2020

See all articles by Andreas Lanz

Andreas Lanz

HEC Paris - Marketing

Gregor Reich

Norwegian School of Economics (NHH) - Department of Strategy and Management

Ole Wilms

Tilburg University - Tilburg University School of Economics and Management

Date Written: September 5, 2020

Abstract

This paper develops a method to flexibly adapt interpolation grids of value function approximations in the estimation of dynamic models using either NFXP (Rust, 1987) or MPEC (Su and Judd, 2012). Since MPEC requires the grid structure for the value function approximation to be hard-coded into the constraints, one cannot apply iterative node insertion for grid refinement; for NFXP, grid adaption by (iteratively) inserting new grid nodes will generally lead to discontinuous likelihood functions. Therefore, we show how to continuously adapt the grid by moving the nodes, a technique referred to as r-adaption. We demonstrate how to obtain optimal grids based on the balanced error principle, and implement this approach by including additional constraints to the likelihood maximization problem. The method is applied to two models: (i) the bus engine replacement model (Rust, 1987), modified to feature a continuous mileage state, and (ii) to a dynamic model of content consumption using original data from SoundCloud, the largest user-generated content network in the music domain.

Keywords: Dynamic discrete choice models, Adaptive Grids, Balanced Errors, Equioscillation

JEL Classification: C25, C63

Suggested Citation

Lanz, Andreas and Reich, Gregor and Wilms, Ole, Adaptive Grids for the Estimation of Dynamic Models (September 5, 2020). HEC Paris Research Paper No. MKG-2020-1383, Available at SSRN: https://ssrn.com/abstract=2650994 or http://dx.doi.org/10.2139/ssrn.2650994

Andreas Lanz

HEC Paris - Marketing ( email )

Paris
France

Gregor Reich

Norwegian School of Economics (NHH) - Department of Strategy and Management ( email )

Breiviksveien 40
N-5045 Bergen
Norway

Ole Wilms (Contact Author)

Tilburg University - Tilburg University School of Economics and Management ( email )

PO Box 90153
Tilburg, 5000 LE Ti
Netherlands

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

Paper statistics

Downloads
102
Abstract Views
716
rank
314,296
PlumX Metrics