Premium
Estimation of Binary Response Models With Endogenous Regressors
Author(s) -
Kang Changhui,
Lee Myoungjae
Publication year - 2014
Publication title -
pacific economic review
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.34
H-Index - 33
eISSN - 1468-0106
pISSN - 1361-374X
DOI - 10.1111/1468-0106.12076
Subject(s) - estimator , instrumental variable , binary number , transformation (genetics) , control function , econometrics , latent variable , estimation , variable (mathematics) , substitution (logic) , mathematics , function (biology) , least squares function approximation , statistics , computer science , control (management) , economics , artificial intelligence , mathematical analysis , biochemistry , chemistry , arithmetic , evolutionary biology , biology , gene , programming language , management
This paper reviews six approaches to binary response ( y 1 ) structural forms with an endogenous regressor y 2 : (i) the two‐stage least squares estimator‐like substitution approach, (ii) the control function approach, (iii) the system reduced‐form approach, (iv) the artificial instrumental regressor approach, (v) the transformed‐response instrumental variable estimator approach and (vi) the classical maximum likelihood estimator approach. The applicability of the six methods differs greatly, depending on whether y 2 is a continuously distributed random variable or a discrete transformation of a latenty 2 * . We conduct a real‐data‐based simulation study, and provide an empirical illustration. Our overall recommendation is using (i) and (ii), as the others have undesirable features such as analytic complexity in (iii), computational difficulty in (iv) and (vi), and poor finite‐sample performance in (v).