Testing Identifying Assumptions in Nonseparable Panel Data Models
نویسنده
چکیده
Recent work on nonparametric identification of average partial effects (APEs) from panel data require restrictions on individual or time heterogeneity. Identifying assumptions under the “generalized first-differencing” category, such as time homogeneity (Chernozhukov, Fernandez-Val, Hahn, and Newey, 2013), have testable equality restrictions on the distribution of the outcome variable. This paper proposes specification tests based on these restrictions. The test statistics are bootstrap-adjusted Kolmogorov-Smirnov and Cramer-von-Mises statistics that are shown to be asymptotically valid and have good finite-sample properties in Monte Carlo simulations. An empirical application illustrates the merits of testing nonparametric identification. JEL: C1, C14, C21, C23, C25
منابع مشابه
Testing Identifying Assumptions in Nonseparable Panel Data Models Supplementary Appendix
(i) βt(xt → xt|Xi = x) = E[Yit|Xi = x′]− E[Yit|Xi = x], (ii) FYiτ |Xi(.|x) = FYiτ |Xi(.|x) Proof (i) follows by the correlated-random-effects assumption and h(x) = h(x′) βt(xt → xt|Xi = x′) = ∫ (ξt(x ′ t, a, u)− ξt(x, a, u))FAi,Uit|Xi(a, u|x ′) = E[Yit|Xi = xt]− ∫ ξt(xt, a, u))dFAi,Uit|Xi(a, u|h(x)) = E[Yit|Xi = x]− E[Yit|Xi = x]. ∗University of California, Davis, One Shields Ave, Davis CA 9561...
متن کاملNonparametric and Semiparametric Panel Econometric Models: Estimation and Testing
This paper gives a selective review on the recent developments of nonparametric and semiparametric panel data models. We focus on the conventional panel data models with one-way error component structure, partially linear panel data models, varying coe¢ cient panel data models, nonparametric panel data models with multi-factor error structure, and nonseparable nonparametric panel data models. F...
متن کاملIdentifying Distributional Effects of Teachers and Peers in Nonseparable Models∗
This paper looks at the effects of teacher and peer characteristics on student achievement in the STAR Project conducted in Tennessee in the late 1980s. As in standard linear models, the proposed approach considers two types of unobservables: school-specific effects and idiosyncratic disturbances. It generalizes previous empirical research by allowing both effects to enter the structural functi...
متن کاملSpatial Correlation Testing for Errors in Panel Data Regression Model
To investigate the spatial error correlation in panel regression models, various statistical hypothesizes and testings have been proposed. This paper, within introduction to spatial panel data regression model, existence of spatial error correlation and random effects is investigated by a joint Lagrange Multiplier test, which simultaneously tests their existence. For this purpose, joint Lagrang...
متن کاملQuantile and Average Effects in Nonseparable Panel Models
This paper gives identification and estimation results for quantile and average effects in nonseparable panel models, when the distribution of period specific disturbances does not vary over time. Bounds are given for interesting effects with discrete regressors that are strictly exogenous or predetermined. We allow for location and scale time effects and show how monotonicity can be used to sh...
متن کامل