Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/142174 
Year of Publication: 
2014
Citation: 
[Journal:] Applied Economics [ISSN:] 1466-4283 [Volume:] 46 [Issue:] 7 [Publisher:] Taylor and Francis [Place:] Milton Park, Abingdon [Year:] 2014 [Pages:] 762-768
Publisher: 
Taylor and Francis, Milton Park, Abingdon
Abstract: 
The classical Heckman (1976, 1979) selection correction estimator (heckit) is misspecified and inconsistent, if an interaction of the outcome variable with an explanatory variable matters for selection. To address this specification problem, a full information maximum likelihood (FIML) estimator and a simple two-step estimator are developed. Monte Carlo (MC) simulations illustrate that the bias of the ordinary heckit estimator is removed by these generalized estimation procedures. Along with OLS and ordinary heckit, we apply these estimators to data from a randomized trial that evaluates the effectiveness of financial incentives for reducing obesity. Estimation results indicate that the choice of the estimation procedure clearly matters.
Subjects: 
selection bias
interaction
heterogeneity
generalized estimator
JEL: 
C24
C93
Published Version’s DOI: 
Document Type: 
Article
Document Version: 
Accepted Manuscript (Postprint)

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.