Abstract
We present a simple way to estimate the effects of changes in a vector of observable variables X on a limited dependent variable Y when Y is a general nonseparable function of X and unobservables, and X is independent of the unobservables. We treat models in which Y is censored from above, below, or both. The basic idea is to first estimate the derivative of the conditional mean of Y given X at x with respect to x on the uncensored sample without correcting for the effect of x on the censored population. We then correct the derivative for the effects of the selection bias. We discuss nonparametric and semiparametric estimators for the derivative. We also discuss the cases of discrete regressors and of endogenous regressors in both cross section and panel data contexts.
Original language  English 

Pages (fromto)  17011719 
Number of pages  19 
Journal  Econometrica 
Volume  80 
Issue number  4 
DOIs 

Publication status  Published  2012 Jul 1 
Externally published  Yes 
Fingerprint
Keywords
 Average derivatives
 Censored dependent variables
 Extreme quantiles
 Nonparametric
 Nonseparable models
 Semiparametric
ASJC Scopus subject areas
 Economics and Econometrics
Cite this
Estimating Derivatives in Nonseparable Models With Limited Dependent Variables. / Altonji, Joseph G.; Ichimura, Hidehiko; Otsu, Taisuke.
In: Econometrica, Vol. 80, No. 4, 01.07.2012, p. 17011719.Research output: Contribution to journal › Comment/debate
}
TY  JOUR
T1  Estimating Derivatives in Nonseparable Models With Limited Dependent Variables
AU  Altonji, Joseph G.
AU  Ichimura, Hidehiko
AU  Otsu, Taisuke
PY  2012/7/1
Y1  2012/7/1
N2  We present a simple way to estimate the effects of changes in a vector of observable variables X on a limited dependent variable Y when Y is a general nonseparable function of X and unobservables, and X is independent of the unobservables. We treat models in which Y is censored from above, below, or both. The basic idea is to first estimate the derivative of the conditional mean of Y given X at x with respect to x on the uncensored sample without correcting for the effect of x on the censored population. We then correct the derivative for the effects of the selection bias. We discuss nonparametric and semiparametric estimators for the derivative. We also discuss the cases of discrete regressors and of endogenous regressors in both cross section and panel data contexts.
AB  We present a simple way to estimate the effects of changes in a vector of observable variables X on a limited dependent variable Y when Y is a general nonseparable function of X and unobservables, and X is independent of the unobservables. We treat models in which Y is censored from above, below, or both. The basic idea is to first estimate the derivative of the conditional mean of Y given X at x with respect to x on the uncensored sample without correcting for the effect of x on the censored population. We then correct the derivative for the effects of the selection bias. We discuss nonparametric and semiparametric estimators for the derivative. We also discuss the cases of discrete regressors and of endogenous regressors in both cross section and panel data contexts.
KW  Average derivatives
KW  Censored dependent variables
KW  Extreme quantiles
KW  Nonparametric
KW  Nonseparable models
KW  Semiparametric
UR  http://www.scopus.com/inward/record.url?scp=84864305014&partnerID=8YFLogxK
UR  http://www.scopus.com/inward/citedby.url?scp=84864305014&partnerID=8YFLogxK
U2  10.3982/ECTA8004
DO  10.3982/ECTA8004
M3  Comment/debate
AN  SCOPUS:84864305014
VL  80
SP  1701
EP  1719
JO  Econometrica
JF  Econometrica
SN  00129682
IS  4
ER 