Selection bias corrections based on the multinomial logit model: Monte Carlo comparisons
François Bourguignon
- Fonction : Auteur
- PersonId : 1014954
- IdHAL : francoisbourguignon
- ORCID : 0000-0002-3094-1827
- IdRef : 026746255
Marc Gurgand
- Fonction : Auteur
- PersonId : 743113
- IdHAL : marc-gurgand
- IdRef : 083172238
Résumé
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.
Domaines
Economies et financesFormat du dépôt | Notice |
---|---|
Type de dépôt | Article dans une revue |
Titre |
en
Selection bias corrections based on the multinomial logit model: Monte Carlo comparisons
|
Résumé |
en
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.
|
Auteur(s) |
François Bourguignon
1, 2, 3
, Marc Gurgand
1, 3, 4
, Martin Fournier
5
1
PSE -
Paris School of Economics
( 301309 )
- 48 boulevard Jourdan 75014 Paris
- France
2
La Banque mondiale - The World Bank
( 51248 )
- 1818 H Street, NW Washington, DC 20433 USA
- États-Unis
3
PJSE -
Paris-Jourdan Sciences Economiques
( 1312 )
- 48 boulevard Jourdan 75014 Paris
- France
4
CREST -
Centre de Recherche en Économie et Statistique
( 2579 )
- 5, Avenue Henry Le Chatelier
91120 Palaiseau
- France
5
GATE -
Groupe d'analyse et de théorie économique
( 785 )
- 93 chemin des Mouilles 69130 ECULLY
- France
|
Comité de lecture |
Oui
|
Vulgarisation |
Non
|
Langue du document |
Anglais
|
Nom de la revue |
|
Audience |
Internationale
|
Date de publication |
2007
|
Volume |
21
|
Numéro |
1
|
Page/Identifiant |
174-205
|
Domaine(s) |
|
Mots-clés |
ro
Selection bias, Multinomial logit, Monte Carlo
|
DOI | 10.1111/j.1467-6419.2007.00503.x |
Loading...