Random Projection Estimation of Discrete-Choice Models with Large Choice Sets

Jan 1, 2019·
Khai Chiong
,
Matt Shum
· 1 min read
Abstract
This paper introduces random projection methods for the estimation of discrete-choice models with large choice sets. By projecting high-dimensional choice features into lower dimensions, the authors achieve efficient estimation without compromising accuracy.
Type
Publication
Management Science, Volume 65, No. 1, pages 256–271, January 2019
publications

Chiong and Shum tackle the computational challenges posed by discrete-choice models with large choice sets. Their random projection estimator dramatically reduces dimensionality, allowing researchers to analyze complex choice data efficiently. The approach is particularly useful for modern marketing and transportation applications involving many alternatives.