Saraswata Chaudhuri

Awareness and the demand for environmental quality: survey evidence on drinking water in urban India

The demand for environmental quality is often presumed to be low in developing countries due to poverty. Less attention has been paid to the possibility that lack of awareness about the adverse health effects of environmental pollution could also keep the demand low. We use a household survey from urban India to estimate the effects of awareness proxies such as schooling and exposure to mass media controlling for wealth on home water purification. Average costs of different home purification methods are used to get estimates on willingness to pay for better drinking water quality. We find that our awareness proxy measures have statistically significant effects on adoption of different home purification methods and therefore, on willingness to pay. These effects are similar in magnitude to the wealth effects.

A New Method of Projection-Based Inference in GMM with Weakly Identified Nuisance Parameters

Projection-based tests for subsets of parameters are useful because they do not over-reject the true parameter values when either it is difficult to estimate the nuisance parameters or their identification status is questionable. However, they are also often criticized for being overly conservative. We overcome this conservativeness by introducing a new projection-based test that is more powerful than the traditional projection-based tests. The new test is even asymptotically equivalent to the related plug-in-based tests when all the parameters are identified. Extension to models with weakly identified parameters shows that the new test is not dominated by the related plug-in-based tests.

A New Projection-Type Split-Sample Score Test In Linear Instrumental Variables Regression

In this paper we introduce a new method of projection-type inference and describe it in the context of two stage least squares–based split-sample inference on subsets of structural coefficients in a linear instrumental variables regression model. The use of the new method not only guards against the uncontrolled overrejection of the true value of the parameters of interest but also reduces the conservativeness of the usual method of projection proposed by Dufour and his coauthors (Dufour, 1997, Econometrica 65, 1365–1388; Dufour and Jasiak, 2001, International Economic Review 41, 815–843; Dufour and Taamouti, 2005, discussion paper; Dufour and Taamouti, 2005, Econometrica 73, 1351–1365; Dufour and Taamouti, 2007, Journal of Econometrics 139, 133–153).

Estimating the Veteran Effect with Endogenous Schooling When Instruments are Potentially Weak

Instrumental variables estimates of the effect of military service on subsequent civilian earnings either omit schooling or treat it as exogenous. In a more general setting that also allows for the treatment of schooling as endogenous, we estimate the veteran effect for men who were born between 1944 and 1952 and thus reached draft age during the Vietnam era. We apply a variety of state-of-the-art econometric techniques to gauge the sensitivity of the estimates to the treatment of schooling. We find a significant veteran penalty.

Shrinkage of Variance for Minimum Distance Based Tests

This paper promotes information theoretic inference in the context of minimum distance estimation. Various score test statistics differ only through the embedded estimator of the variance of estimating functions. We resort to implied probabilities provided by the constrained maximization of generalized entropy to get a more accurate variance estimator under the null. We document, both by theoretical higher order expansions and by Monte-Carlo evidence, that our improved score tests have better finite-sample size properties. The competitiveness of our non-simulation based method with respect to bootstrap is confirmed in the example of inference on covariance structures previously studied by Horowitz (1998).

Projection-Based Gel Score Test for Subsets of Parameters with Possible Weak Identification

It has been recently shown that generalized empirical likelihood (GEL) methods can be used to design score tests for subsets of parameters such that the asymptotic size of the tests is equal to their nominal size when the nuisance parameters in the model are (strongly) identified. However, this does not necessarily hold if the nuisance parameters are not identified; and in such cases there is no guarantee that the standard (first-order) asymptotic chi-squared pproximation of the score statistic will not result in over-rejection of the true value of the parameters of interest. In this paper we address this problem by proposing a new method of projection-based score test that guards against the uncontrolled over-rejection of the true value of the parameters of interest even when the nuisance parameters are not identified, while achieving asymptotic equivalence with the GEL score tests otherwise.