Working papers

We propose strategies to estimate and make inference on key features of heterogeneous effects in randomized experiments.

The R package Counterfactual implements the estimation and inference methods of Chernozhukov et al. (2013) for counterfactual analysis.

This paper introduces new inference methods for counterfactual and synthetic control methods for evaluating policy effects.

We look at how strongly shocks to asset values affect labour supply, using Italian data.

High-dimensional linear models with endogenous variables play an increasingly important role in recent econometric literature.

This paper considers nonparametric additive models that have a deterministic time trend and both stationary and integrated variables as components.

CEPR Discussion Paper DP12499

This paper proposes a specification test of the mixed logit models, by generalizing Hausman and McFadden’s (1984) test. We generalize the test even further by considering a model developed by Berry, Levinsohn and Pakes (1995).

This paper proposes a powerful alternative to the t-test in linear regressions when a regressor is mismeasured.

This paper develops and implements a nonparametric test of Random Utility Models. The motivating application is to test the null hypothesis that a sample of cross-sectional demand distributions was generated by a population of rational consumers.

We study optimal corrective taxation in the alcohol market. Consumption generates negative externalities that are non-linear in the total amount of alcohol consumed. If tastes for products are heterogeneous and correlated with marginal externalities, then varying tax rates on different products can lead to welfare gains.

We derive testable implications of model in which first best allocations are not achieved because of a moral hazard problem with hidden saving.

This paper examines the trends in geographic localization of knowledge spillovers via patent citations, considering US patents from the period of 1976-2015.

The understanding of co-movements, dependence, and influence between variables of interest is key in many applications. Broadly speaking such understanding can lead to better predictions and decision making in many settings.

Soda taxes aim to reduce excessive sugar consumption. Their effectiveness depends on whether they target individuals for whom the harm of consumption is largest. We estimate demand and account for supply-side equilibrium pass-through. We exploit longitudinal data to estimate individual preferences, which allows exible heterogeneity that we relate to a wide array of individual characteristics. We show that soda taxes are effective at targeting young consumers but not individuals with high total dietary sugar; they impose the highest monetary cost on poorer individuals, but are unlikely to be strongly regressive if we account for averted future costs from over consumption.

We propose a new estimator for the dynamic panel model, which solves the failure of strict exogeneity by calculating the bias in the first-order conditions as a function of the autoregressive parameter and solving the resulting equation.

This paper estimates the effects of entering the labour market when the economy is weak on subsequent living standards using consistent long-running household survey data from the UK.

This paper studies inference of preference parameters in semiparametric discrete choice models when these parameters are not point-identified and the identified set is characterized by a class of conditional moment inequalities.

We consider a variable selection problem for the prediction of binary outcomes. We study the best subset selection procedure by which the explanatory variables are chosen by maximizing Manski (1975, 1985)'s maximum score type objective function subject to a constraint on the maximal number of selected variables.

We show that the generalized method of moments (GMM) estimation problem in instrumental variable quantile regression (IVQR) models can be equivalently formulated as a mixed integer quadratic programming problem.