Premium
SELECTION BIAS CORRECTIONS BASED ON THE MULTINOMIAL LOGIT MODEL: MONTE CARLO COMPARISONS
Author(s) -
Bourguig François,
Fournier Martin,
Gurgand Marc
Publication year - 2007
Publication title -
journal of economic surveys
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.657
H-Index - 92
eISSN - 1467-6419
pISSN - 0950-0804
DOI - 10.1111/j.1467-6419.2007.00503.x
Subject(s) - monte carlo method , multinomial logistic regression , econometrics , estimator , selection (genetic algorithm) , logit , parametric statistics , model selection , multinomial distribution , selection bias , set (abstract data type) , statistics , mathematics , computer science , artificial intelligence , programming language
This survey presents the set of methods available in the literature on selection bias correction, when selection is specified as a multinomial logit model. It contrasts the underlying assumptions made by the different methods and shows results from a set of Monte Carlo experiments. We find that, in many cases, the approach initiated by Dubin and MacFadden (1984) as well as the semi‐parametric alternative recently proposed by Dahl (2002) are to be preferred to the most commonly used Lee (1983) method. We also find that a restriction imposed in the original Dubin and MacFadden paper can be waived to achieve more robust estimators. Monte Carlo experiments also show that selection bias correction based on the multinomial logit model can provide fairly good correction for the outcome equation, even when the IIA hypothesis is violated.