A Monte Carlo Study of Tests for the Independence of Irrelevant Alternatives Property

A plethora of tests for the Independence of Irrelevant Alternatives (IA) property of Logit models of discrete choice behavior has been proposed in the literature. These tests are based upon asymptotic arguments and little is known about their size and power properties in finite samples. This paper uses a Monte Carlo simulation study to investigate the size and power properties of six tests for IIA in the multinomial Logit (MNL) model. Our results show that tests based upon partitioning the choice set appear to have very poor size and power properties in small samples.


Issue Date:
Sep 01 1993
Publication Type:
Working or Discussion Paper
Language:
English
Total Pages:
24
JEL Codes:
C25; C12
Series Statement:
Working Paper No. 3/93




 Record created 2018-01-30, last modified 2018-01-31

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)