What can we learn about correlations from multinomial probit estimates?

Monfardini, Chiara and Santos Silva, J.M.C. (2006) What can we learn about correlations from multinomial probit estimates? Bologna: Dipartimento di Scienze economiche DSE, p. 17. DOI 10.6092/unibo/amsacta/4729. In: Quaderni - Working Paper DSE (558). ISSN 2282-6483.
Full text available as:
[img]
Preview
Text(pdf)
License: Creative Commons Attribution

Download (273kB) | Preview

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.

Abstract
Document type
Monograph (Working Paper)
Creators
CreatorsAffiliationORCID
Monfardini, Chiara
Santos Silva, J.M.C.
Keywords
Correlations; Equicorrelation; Identification; Inequality restrictions
Subjects
ISSN
2282-6483
DOI
Deposit date
29 Feb 2016 11:39
Last modified
29 Feb 2016 11:39
URI

Other metadata

Downloads

Downloads

Staff only: View the document

^