Testing the Information Matrix Equality with Robust Estimators



Testing the Information Matrix Equality
with Robust Estimators

Christophe Croux
K.U.Leuven


Geert Dhaene
K.U.Leuven


Dirk Hoorelbeke *
K.U.Leuven


March 18, 2003

Abstract

We study the behaviour of the information matrix (IM) test when
maximum likelihood estimators are replaced with robust estimators.
The latter may unmask outliers and hence improve the power of the
test. We investigate in detail the local asymptotic power of the IM
test in the normal model, for various estimators and under a range
of local alternatives. These local alternatives include contamination
neighbourhoods, Student’s
t (with degrees of freedom approaching in-
finity), skewness, and a tilted normal. Simulation studies for fixed
alternatives confirm that in many cases the use of robust estimators
substantially increases the power of the IM test.

JEL classification: C12, C15

Key-words: information matrix test, robustness

* Corresponding author. Address: K.U.Leuven, Department of Economics, Naamses-
traat 69, 3000 Leuven, Belgium. Tel. +32 16 326652. Fax +32 16 326796. Email:
[email protected]



More intriguing information

1. School Effectiveness in Developing Countries - A Summary of the Research Evidence
2. The mental map of Dutch entrepreneurs. Changes in the subjective rating of locations in the Netherlands, 1983-1993-2003
3. The name is absent
4. The name is absent
5. Epistemology and conceptual resources for the development of learning technologies
6. The name is absent
7. Making International Human Rights Protection More Effective: A Rational-Choice Approach to the Effectiveness of Ius Standi Provisions
8. The name is absent
9. Dual Track Reforms: With and Without Losers
10. LOCAL PROGRAMS AND ACTIVITIES TO HELP FARM PEOPLE ADJUST