What Can We Learn About Correlations from Multinomial Probit Estimates?

19 Pages Posted: 28 Feb 2006

See all articles by Chiara Monfardini

Chiara Monfardini

University of Bologna - Department of Economics; IZA Institute of Labor Economics

J. M.C. Santos Silva

University of Surrey

Date Written: May 2006

Abstract

It is well known that, in a multinomial probit, only the covariance matrix of the location and scale normalized utilities are identified. In this study, we explore the relation between these identifiable parameters and the original elements of the covariance matrix, to find out what can be learnt about the correlations between the stochastic components of the non-normalized utilities. We show that, in certain circumstances, it is possible to obtain information on these behavioural parameters and define appropriate tools for inference. We illustrate the usefulness of our results in applied settings using an example.

Keywords: Correlations, Equicorrelation, Identification, Inequality restrictions

JEL Classification: C25

Suggested Citation

Monfardini, Chiara and Santos Silva, João M.C, What Can We Learn About Correlations from Multinomial Probit Estimates? (May 2006). Available at SSRN: https://ssrn.com/abstract=886495 or http://dx.doi.org/10.2139/ssrn.886495

Chiara Monfardini (Contact Author)

University of Bologna - Department of Economics ( email )

Piazza Scaravilli 2
Bologna, 40126
Italy
0039 51 2098148 (Phone)
0039 51 221968 (Fax)

IZA Institute of Labor Economics

P.O. Box 7240
Bonn, D-53072
Germany

João M.C Santos Silva

University of Surrey ( email )

Guildford
Surrey GU2 7XH
United Kingdom

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

Paper statistics

Downloads
89
Abstract Views
770
rank
354,581
PlumX Metrics