Results 1  10
of
121
Dynamic Discrete Choice and Dynamic Treatment Effects
, 2005
"... This paper considers semiparametric identification of structural dynamic discrete choice models and models for dynamic treatment effects. Time to treatment and counterfactual outcomes associated with treatment times are jointly analyzed. We examine the implicit assumptions of the dynamic treatment m ..."
Abstract

Cited by 135 (30 self)
 Add to MetaCart
This paper considers semiparametric identification of structural dynamic discrete choice models and models for dynamic treatment effects. Time to treatment and counterfactual outcomes associated with treatment times are jointly analyzed. We examine the implicit assumptions of the dynamic treatment model using the structural model as a benchmark. For the structural model we show the gains from using cross equation restrictions connecting choices to associated measurements and outcomes. In the dynamic discrete choice model, we identify both subjective and objective outcomes, distinguishing ex post and ex ante outcomes. We show how to identify agent information sets.
Endogeneity in Nonparametric and Semiparametric Regression Models
, 2000
"... 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 — t ..."
Abstract

Cited by 130 (19 self)
 Add to MetaCart
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 be applicable to nonparametric generalizations of the linear model and to their semiparametric variants. The discussion then turns to a particular semiparametric model, the binary response model with linear index function and nonparametric error distribution, and describes in detail how estimation of the parameters of interest can be constructed using the “control function ” approach. This estimator is then applied to an empirical problem of the relation of labor force participation to nonlabor income, viewed as an endogenous regressor.
Estimation of Nonparametric Simultaneous Equations
, 2005
"... This paper considers identification in parametric and nonparametric models, with additive or nonadditive unobservables, and with or without simultaneity among the endogenous variables. Several characterizations of observational equivalence are presented and conditions for identification are develope ..."
Abstract

Cited by 40 (7 self)
 Add to MetaCart
This paper considers identification in parametric and nonparametric models, with additive or nonadditive unobservables, and with or without simultaneity among the endogenous variables. Several characterizations of observational equivalence are presented and conditions for identification are developed based on these. It is shown that the results can be extended to situations where the dependent variables are latent. We also demonstrate how the results may be used to derive constructive ways to calculate the unknown functions and distributions in simultaneous equations models, directly from the probability density of the observable variables. Estimators based on this do not suffer from the illposed inverse problem that other methods encounter.
Semiparametric Binary Choice Panel Data Models Without Strictly Exogeneous Regressors,” unpublished manuscript
, 2000
"... Most previous studies of binary choice panel data models with Þxed effects require strictly exogeneous regressors, and except for the logit model without lagged dependent variables, cannot provide rate root n parameter estimates. We assume that one of the explanatory variables is independent of the ..."
Abstract

Cited by 35 (6 self)
 Add to MetaCart
Most previous studies of binary choice panel data models with Þxed effects require strictly exogeneous regressors, and except for the logit model without lagged dependent variables, cannot provide rate root n parameter estimates. We assume that one of the explanatory variables is independent of the individual speciÞc effect and of the errors of the model, conditional on the other explanatory variables. Based on Lewbel (2000a), we show how this alternativeassumptioncanbeusedtoidentifyandrootn consistently estimate the parameters of discrete choice panel data models with Þxed effects, only requiring predetermined (as opposed to strictly exogeneous) regressors. The estimator is semiparametric in that the error distribution is not speciÞed, and allows for general forms of heteroscedasticity.
Nonparametric Identification of Multinomial Choice Demand Models with Heterogeneous Consumers
, 2008
"... We consider identification of nonparametric random utility models of multinomial choice using observation of consumer choices. Our model of preferences nests random coefficients discrete choice models widely used in practice with parametric functional form and distributional assumptions. However, ou ..."
Abstract

Cited by 33 (0 self)
 Add to MetaCart
We consider identification of nonparametric random utility models of multinomial choice using observation of consumer choices. Our model of preferences nests random coefficients discrete choice models widely used in practice with parametric functional form and distributional assumptions. However, our model is nonparametric and distribution free. It incorporates choicespecific unobservables, endogenous choice characteristics, unknown heteroskedasticity, and correlated taste shocks. We consider full identification of the random utility model as well as identification of demand. Under standard orthogonality, “large support,” and instrumental variables assumptions, we show identifiability of choicespecific unobservables and the joint distribution of preferences conditional on any vector of observed and unobserved characteristics. We demonstrate robustness of these results to relaxation of the large support condition and show that when this condition is replaced with a much weaker “common choice probability” condition, the demand structure is still identified. We also show that key maintained hypotheses are testable. We have had helpful conversations on this topic with Hide Ichimura, Rosa Matzkin and Yuichi Kitamura. We
Estimating Features of a Distribution From Binomial Data
 Journal of Econometrics
"... A statistical problem that arises in several fields is that of estimating the features of an unknown distribution, which may be conditioned on covariates, using a sample of binomial observations on whether draws from this distribution exceed threshold levels set by experimental design. Applications ..."
Abstract

Cited by 28 (18 self)
 Add to MetaCart
A statistical problem that arises in several fields is that of estimating the features of an unknown distribution, which may be conditioned on covariates, using a sample of binomial observations on whether draws from this distribution exceed threshold levels set by experimental design. Applications include bioassay and destructive duration analysis. The empirical application we consider is referendum contingent valuation in resource economics, where one is interested in features of the distribution of values (willingness to pay) placed by consumers on a public good such as endangered species. Sample consumers are asked whether they favor a referendum that would provide the good at a cost specified by experimental design. This paper provides estimators for moments and quantiles of the unknown distribution in this problem under both nonparametric and semiparametric specifications.
Endogenous Selection or Treatment Model Estimation
 Journal of Econometrics
, 2007
"... In a sample selection or treatment effects model, common unobservables may affect both the outcome and the probability of selection in unknown ways. This paper shows that the distribution function of potential outcomes, conditional on covariates, can be identifiedgivenanobservedvariableVthat affects ..."
Abstract

Cited by 27 (7 self)
 Add to MetaCart
In a sample selection or treatment effects model, common unobservables may affect both the outcome and the probability of selection in unknown ways. This paper shows that the distribution function of potential outcomes, conditional on covariates, can be identifiedgivenanobservedvariableVthat affects the treatment or selection probability in certain ways and is conditionally independent of the potential outcome equation error terms. Selection model estimators based on this identification are provided, which take the form of either simple weighted averages or GMM or two stage least squares. These estimators permit endogenous and mismeasured regressors. Empirical applications are provided to estimation of a firm investment model and a returns to schooling wage model. Portions of this paper were previously circulated under other titles including, ”Two Stage Least Squares Estimation
Semiparametric estimation of multinomial discretechoice models using a subset of choices
 RAND JOURNAL OF ECONOMICS
, 2007
"... ..."
Identification in Differentiated Products Markets Using Market Level Data
, 2009
"... We consider nonparametric identification in models of differentiated products markets, using only market level observables. On the demand side we consider a nonparametric random utility model nesting random coefficients discrete choice models widely used in applied work. We allow for product/market ..."
Abstract

Cited by 26 (2 self)
 Add to MetaCart
We consider nonparametric identification in models of differentiated products markets, using only market level observables. On the demand side we consider a nonparametric random utility model nesting random coefficients discrete choice models widely used in applied work. We allow for product/marketspecific unobservables, endogenous product characteristics (e.g., prices), and highdimensional taste shocks with arbitrary correlation and heteroskedasticity. On the supply side we specify marginal costs nonparametrically, allow for unobserved firm heterogeneity, and nest a variety of equilibrium oligopoly models. We pursue two approaches to identification. One relies on instrumental variables conditions used previously to demonstrate identification in a nonparametric regression framework. With this approach we can show identification of the demand side without reference to a particular supply model. Adding the supply side allows identification of firms’ marginal costs as well. Our second approach, more closely linked to classical identification arguments for supply and demand models, employs a change of variables approach. This leads to constructive identification results relying on exclusion and support conditions. Our results lead to a testable restriction that provides the first general formalization of Bresnahan’s (1982) intuition for empirically discriminating between alternative models of oligopoly competition.