A Multinomial Probit Model with Latent Factors: Identification and Interpretation Without a Measurement System

49 Pages Posted: 2 Oct 2017

See all articles by Remi Piatek

Remi Piatek

University of Chicago

Miriam Gensowski

University of Copenhagen

Abstract

We develop a parametrization of the multinomial probit model that yields greater insight into the underlying decision-making process, by decomposing the error terms of the utilities into latent factors and noise. The latent factors are identified without a measurement system, and they can be meaningfully linked to an economic model. We provide sufficient conditions that make this structure identified and interpretable. For inference, we design a Markov chain Monte Carlo sampler based on marginal data augmentation. A simulation exercise shows the good numerical performance of our sampler and reveals the practical importance of alternative identification restrictions. Our approach can generally be applied to any setting where researchers can specify an a priori structure on a few drivers of unobserved heterogeneity. One such example is the choice of combinations of two options, which we explore with real data on education and occupation pairs.

Keywords: multinomial probit, latent factors, Bayesian analysis, marginal data augmentation, educational choice, occupational choice

JEL Classification: C11, C25, C35

Suggested Citation

Piatek, Rémi and Gensowski, Miriam, A Multinomial Probit Model with Latent Factors: Identification and Interpretation Without a Measurement System. IZA Discussion Paper No. 11042, Available at SSRN: https://ssrn.com/abstract=3045739 or http://dx.doi.org/10.2139/ssrn.3045739

Rémi Piatek (Contact Author)

University of Chicago ( email )

1101 East 58th Street
Chicago, IL 60637
United States

Miriam Gensowski

University of Copenhagen ( email )

Nørregade 10
Copenhagen, København DK-1165
Denmark

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

Paper statistics

Downloads
26
Abstract Views
367
PlumX Metrics