This fall I will join the Economics Department at the University of Pennsylvania.

Education

Research Interests

Publications

Measuring Altruism in a Public Goods Experiment: A Comparison of U.S. and Czech Subjects, (with Lisa Anderson and Jeffrey Gerlach), Experimental Economics, 2011, 14 (3), pp. 426-437.

Yes, Wall Street, There is a January Effect: Evidence from Laboratory Auctions, (with Lisa Anderson and Jeffrey Gerlach), Journal of Behavioral Finance, 2007, 8 (1), pp. 1-8.

Job Market Paper

Using Invalid Instruments on Purpose: Focused Moment Selection and Averaging for GMM

In finite samples, the use of a slightly invalid but highly relevant instrument can substantially reduce mean-squared error (MSE). Building on this observation, I propose a moment selection criterion for GMM in which over-identifying restrictions are chosen based on the MSE of their associated estimators rather than their validity: the focused moment selection criterion (FMSC). I then show how the asymptotic framework used to derive the FMSC can be employed to address the problem of inference post-moment selection. Treating post-selection estimators as a special case of moment-averaging, in which estimators based on different moment sets are given data-dependent weights, I propose a simulation-based procedure to construct valid confidence intervals. In a Monte Carlo experiment for 2SLS estimation, the FMSC performs well relative to alternatives suggested in the literature, and the simulation-based procedure achieves its stated minimum coverage. I conclude with an empirical example examining the effect of instrument selection on the estimated relationship between malaria transmission and economic development.

Working Papers

Portfolio Selection: An Extreme Value Approach (with Jeffrey Gerlach), Revise and Resubmit Journal of Banking and Finance

We show that lower tail dependence (chi), a measure of the probability that a portfolio will suffer large losses given that the market does, contains important information for risk-averse investors. We then estimate chi for a sample of DJIA stocks and show that it differs systematically from other risk measures including variance, semi-variance, skewness, kurtosis, beta, and coskewness. In out-of-sample tests, portfolios constructed to have low values of chi outperform the market index, the mean return of the stocks in our sample, and portfolios with high values of chi. Our results indicate that chi is conceptually important for risk-averse investors and provides useful information for portfolio selection.

Research in Progress

A Generalized Focused Information Criterion with Application to Dynamic Panel Models

Avoiding Moment Selection with Bayesian Nonparametrics

This paper suggests a Bayesian solution to the many instruments problem for linear IV models identified by conditional moment restrictions. Rather than attempting to select the most important functions of the underlying instruments for use in estimation, we include them all. Placing a Gaussian Process prior on the conditional mean function of the endogenous regressor yields a computationally tractable, data-driven shrinkage estimator. Although the prior is infinite-dimensional, the model is conditionally conjugate: we show that draws from the posterior of the parameters of interest can be obtained from a straightforward Gibbs Sampler. By placing a hierarchical prior on the parameters of the Gaussian Process, we further exploit conditional conjugacy to combine prior uncertainty about the smoothness of the conditional mean function with relevant information contained in the data themselves. This provides an important advantage over frequentist approaches in which smoothing parameters, even if chosen on the basis of the data, are ultimately treated as fixed.

Faculty of Economics
Austin Robinson Building
Sidgwick Avenue
Cambridge, CB3 9DD
United Kingdom

Tel: +44 (0)7859 415 520
Email: fjd26@cam.ac.uk