Research

Research Interest

• Partial Identification
• Optimal Selection of Instruments
• Quantile Regression
• Empirical Likelihood

Working Papers

1. EL Inference for Partially Identified Models: Large Deviations Optimality and Bootstrap Validity
   Revised:   February 2008 with Supplementary Appendix
   
   Abstract (show/hide)

   This paper addresses the issue of optimal inference for parameters that are partially identified in models with moment inequalities. There currently exists a variety of inferential methods for use in this setting. However, the question of choosing optimally among contending procedures is unresolved. In this paper, I first consider two canonical large deviations criteria for optimality and show that inference based on the empirical likelihood ratio statistic is optimal according to both criteria. This finding is a direct analog to the findings in Kitamura (2001) and Kitamura and Otsu (2005) for moment equality models. Second, I introduce a simple and natural modification of the empirical likelihood bootstrap by Brown and Newey (2002) which provides a valid bootstrap method for moment inequality models. This procedure overcomes the implementation challenges that arise in partially identified models as a result of non-pivotal limit distributions and the fact that the standard empirical likelihood bootstrap is inconsistent. Lastly, I analyze the finite sample properties of the proposed framework using Monte Carlo simulations. The simulation results are encouraging.

2. Simultaneous Selection and Weighting of moments in GMM using a Trapezoidal Kernel
   Revised:   March 2008.
   
   Abstract (show/hide)

   This paper proposes a novel procedure to estimate linear models when the number of instruments is large. At the heart of such models is the need to balance the trade off between attaining asymptotic efficiency, which requires more instruments, and minimizing bias, which is adversely affected by the addition of instruments. In fact, two questions are of central concern: (1) What is the optimal number of instruments to use? (2) Should the instruments receive different weights? This paper contains the following contributions toward resolving these issues. First, I propose a kernel weighted generalized method of moments estimator that uses a trapezoidal kernel. This kernel turns out to be attractive to select and weight the number of moments. Second, I derive the higher order mean squared error of the kernel weighted GMM estimator and show that the trapezoidal kernel generates a lower asymptotic variance than regular kernels. The approach is similar in spirit to Kuersteiner(2002), though the derivation and the resulting expansion differ substantially due to particular features of the trapezoidal kernel. Third, I suggest the bootstrap as a feasible data-dependent bandwidth selection rule. Finally, I analyze the finite sample properties of the estimator using Monte Carlo simulations. The results show that the kernel weighted GMM estimator performs on par with other estimators that choose optimal instruments and improves upon a GMM estimator that uses all instruments.

3. A Note on Quantile Regression for Panel Data Models
   This Version:   April 2008.
   
   Abstract (show/hide)

   Despite the widespread use of both panel data models and quantile regression models, there has been little work at the intersection of these two methodologies. This fact is possibly due to two fundamental issues associated with conditional quantiles. First, standard demeaning techniques do not result in feasible approaches. Second, the interpretation typically given to individual effects is less clear in quantile regression models. This note addresses the first issue by proposing a simple transformation of the data that gets ride of the fixed effects under the assumption that these effects are location shifters. The resulting two-step estimator is consistent and asymptotically normal as both n and T grow.

Old Papers

• A Bootstrap Approach to Bandwidth Selection in HAC Covariance Matrix Estimation
• Corruption, Ownership and Mixtures of Efficiency Measures: An Application to the Electricity Sector in Latin America.

UW Beamer Theme - LaTeX

I did a UW Style file for Beamer. You are welcome to use it if you like it. What you have to do is: (1) Download the zip file; (2)Unzip the content of the zip file in your folder of choice; (3) READ Example.pdf; (4) Edit Example.tex with your presentation. I use WinEdt so I am not sure whether this file would work with SciWord. I hope you enjoy it!

• Zip File
• PDF Example