NONPARAMETRIC TWO–STEP REGRESSION ESTIMATION WHEN REGRESSORS AND ERROR ARE DEPENDENT Running Head: Nonparametric Two Step Estimation
نویسندگان
چکیده
This paper considers estimation of the function g in the model Yt = g(Xt) + εt when E(εt|Xt) 6= 0 with non–zero probability. We assume the existence of an ‘instrumental variable’ Zt that is independent of εt and of an ‘innovation’ ηt = Xt − E(Xt|Zt). We use a nonparametric regression of Xt on Zt to obtain residuals η̂t which in turn are used to obtain a consistent estimator of g. The estimator was first analyzed by Newey, Powell and Vella (1995) under the assumption that the observations are i.i.d.. Here we derive a sample mean square error convergence result for independent and identically distributed observations as well as a uniform convergence result under time series dependence. Cet article concerne l’estimation de la fonction g dans le modèle Yt = g(Xt) + εt où E[εt|Xt] 6= 0 avec probabilité non nulle. Les auteurs supposent l’existence d’une ‘variable instrumentale’ Zt qui est indépendante de εt et de l’innovation ηt = Xt − E[Xt|Zt]. Les résidus η̂t déduits d’une régression non paramétrique de Xt sur Zt permettent d’obtenir une estimation convergente de g. Cette façon de procéder avait déjà été proposée par Newey, Powell et Vella (1995) dans le cas où les observations forment un échantillon aléatoire. Les auteurs démontrent ici la convergence de l’erreur quadratique moyenne expérimentale sous les mêmes conditions et établissent un résultat de convergence uniforme sous des conditions de dépendance sérielle entre les observations.
منابع مشابه
Instrumental Variables Estimation of Nonparametric Models with Discrete Endogenous Regressors
This paper presents new instrumental variables estimators for nonparametric models with discrete endogenous regressors. The model speci cation is su ciently general to include structural models, triangular simultaneous equations and certain models of measurement error. Restricting the analysis to discrete endogenous regressors is an integral component of the analysis since a similar model with ...
متن کاملBayesian Bandwidth Selection for a Nonparametric Regression Model with Mixed Types of Regressors
This paper develops a sampling algorithm for bandwidth estimation in a nonparametric regression model with continuous and discrete regressors under an unknown error density. The error density is approximated by the kernel density estimator of the unobserved errors, while the regression function is estimated using the Nadaraya-Watson estimator admitting continuous and discrete regressors. We der...
متن کاملRegressions with Berkson errors in covariates – A nonparametric approach
This paper establishes that so-called instrumental variables enable the identification and the estimation of a fully nonparametric regression model with Berkson-type measurement error in the regressors. An estimator is proposed and proven to be consistent. Its practical performance and feasibility are investigated via Monte Carlo simulations as well as through an epidemiological application inv...
متن کاملEndogeneity in Nonparametric and Semiparametric Regression Models
This paper considers the nonparametric and semiparametric methods for estimating regression models with continuous endogenous regressors. We list a number of different generalizations of the linear structural equation model, and discuss how three common estimation approaches for linear equations — the “instrumental variables,” “fitted value,” and “control function” approaches — may or may not b...
متن کاملNonparametric Regression Applied to Quantitative Structure-Activity Relationships
Several nonparametric regressors have been applied to modeling quantitative structure-activity relationship (QSAR) data. The simplest regressor, the Nadaraya-Watson, was assessed in a genuine multivariate setting. Other regressors, the local linear and the shifted Nadaraya-Watson, were implemented within additive models--a computationally more expedient approach, better suited for low-density d...
متن کامل