Comparing IV with structural models: What simple IV can and cannot identify

被引:112
作者
Heckman, James J. [1 ,2 ,3 ]
Urzua, Sergio [4 ]
机构
[1] Univ Chicago, Dept Econ, Chicago, IL 60637 USA
[2] Univ Coll Dublin, Dublin, Ireland
[3] Yale Univ, Cowles Fdn, New Haven, CT 06520 USA
[4] Northwestern Univ, Evanston, IL 60208 USA
关键词
Instrumental variables; Structural approaches; Marschak's Maxim; INSTRUMENTAL VARIABLES; CHOICE; IDENTIFICATION; UNCERTAINTY; ABILITIES; EQUATIONS;
D O I
10.1016/j.jeconom.2009.09.006
中图分类号
F [经济];
学科分类号
02 ;
摘要
This paper compares the economic questions addressed by instrumental variables estimators with those addressed by structural approaches. We discuss Marschak's Maxim: estimators should be selected on the basis of their ability to answer well-posed economic problems with minimal assumptions. A key identifying assumption that allows structural methods to be more informative than IV can be tested with data and does not have to be imposed. (C) 2009 Elsevier B.V. All rights reserved.
引用
收藏
页码:27 / 37
页数:11
相关论文
共 43 条
[21]   The effects of cognitive and noncognitive abilities on labor market outcomes and social behavior [J].
Heckman, James J. ;
Stixrud, Jora ;
Urzua, Sergio .
JOURNAL OF LABOR ECONOMICS, 2006, 24 (03) :411-482
[22]  
Heckman JJ, 2007, HBK ECON, V2, P4875, DOI 10.1016/S1573-4412(07)06071-0
[23]  
Heckman JamesJ., 1987, NEW PALGRAVE, P287
[24]   Structural equations, treatment effects, and econometric policy evaluation [J].
Heckman, JJ ;
Vytlacil, E .
ECONOMETRICA, 2005, 73 (03) :669-738
[25]  
Heckman JJ, 2001, INT SYMP EC, P1
[26]   Local instrumental variables and latent variable models for identifying and bounding treatment effects [J].
Heckman, JJ ;
Vytlacil, EJ .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 1999, 96 (08) :4730-4734
[27]  
HECKMAN JJ, 1992, EVALUATING WELFARE AND TRAINING PROGRAMS, P201
[28]  
HECKMAN JJ, 2008, INSTRUMENTA IN PRESS
[29]  
HECKMAN JJ, 2009, J ECONOMETR IN PRESS
[30]  
HECKMAN JJ, 2008, 14044 NAT BUR EC