Observational studies are widely used to evaluate the effect of treatment when it is not feasible to conduct controlled experiment. This article considers the use of parametric analyses for estimating the causal treatment effect. The proposed approach is an alternative to the widely used...

Across the social sciences, lagged explanatory variables are a common strategy to confront challenges to causal identification using observational data. We show that "lag identification"--the use of lagged explanatory variables to solve endogeneity problems--is an illusion: lagging independent...

In this paper, we study partial identification of the distribution of treatment effects of a binary treatment for ideal randomized experiments, ideal randomized experiments with a known value of a dependence measure, and for data satisfying the selection-on-observables assumption respectively....

We examine challenges to estimation and inference when the objects of interest are nondifferentiable functionals of the underlying data distribution. This situation arises in a number of applications of bounds analysis and moment inequality models, and in recent work on estimating optimal...

I reconsider various methods for correcting for bias in estimates of the returns to schooling. I argue that the literature on ability bias has ignored complications implicit in theoretical formulations of the choice of human capital. In particular, such models imply that adding ability to the...

IV estimators with an instrument vector composed only of past squared residuals, while applicable to the semi-strong ARCH(1) model, do not extend to the semi-strong GARCH(1,1) case because of underidentification. Augmenting the instrument vector with past residuals, however, renders traditional...

This paper gives a new jackknife estimator for instrumental variable inference with unknown heteroskedasticity. The estimator is derived by using a method of moments approach similar to the one that produces LIML in case of homoskedasticity. The estimator is symmetric in the endogenous variables...

We show in this paper that the treatment of conditional heteroskedasticity inside nonlinear systems of simultaneous equations is a sufficiently manageable matter for some types of multivariate ARCH error structures. Reparameterization makes it possible to estimate the model by means of the...

In this paper, we extend Bai and Perron’s (1998, Econometrica, p.47-78) framework for multiple break testing to linear models estimated via Two Stage Least Squares (2SLS). Within our framework, the break points are estimated simultaneously with the regression parameters via minimization of the...

In this paper we construct thirteen different types of composite indices by linear combination of indicator variables (with and without outliers/data corruption). Weights of different indicator variables are obtained by maximization of the sum of squared (and, alternatively, absolute)...