Working Papers
Quantile Effects in a Sample Selection Model for Network and Panel Data, Job Market Paper. November 2018.
Abstract
I develop a distribution regression model with sample selection for panel and network data. The model specifies a bivariate Gaussian distribution of the latent selection and outcome variables semiparametrically with function-valued parameters and unobserved effects. The unobserved effects are included in the selection equation, outcome equation and selection sorting, allowing for very rich patterns of unobserved heterogeneity. I provide a two-step fixed-effect method to estimate the model parameters and other functionals such as distributions and quantile effects. In addition, I derive analytical and Jackknife bias corrections to deal with the incidental parameter problem of the fixed-effect estimators. A multiplier bootstrap algorithm is adopted to construct confidence bands for uniform inference on the model parameters and functionals of interest. I apply this model to the gravity equation of trade network data between countries accounting for possibly endogenous zero trade decisions and unobserved country heterogeneity. The model credibly identifies positive and homogeneous effects of having a common legal system and negative effects of increasing pairwise distance on the latent trade volume that are heterogeneous across the distribution.
Distribution Regression with Sample Selection, with an Application to Wage Decompositions in the UK, with Victor Chernozhukov and Ivan Fernandez-Val. November 2018.
Abstract
We develop a distribution regression model under endogenous sample selection. This model is a semiparametric generalization of the Heckman selection model that accommodates much rich patterns of heterogeneity in the selection process and effect of the covariates. The model applies to continuous, discrete and mixed outcomes. We study the identification of the model, and develop a computationally attractive two-step method to estimate the model parameters, where the first step is a probit regression for the selection equation and the second step consists of multiple distribution regressions with selection corrections for the outcome equation. We construct estimators of functionals of interest such as actual and counterfactual distributions of latent and observed outcomes via plug-in rule. We derive functional central limit theorems for all the estimators and show the validity of multiplier bootstrap to carry out functional inference. We apply the methods to wage decompositions in the UK using new data. Here we decompose the difference between the male and female wage distributions into four effects: composition, wage structure, selection structure and selection sorting. We uncover positive sorting for single men and negative sorting for married women that accounts for a substantial fraction of the gender wage gap at the top of the distribution. These findings can be interpreted as evidence of assortative matching in the marriage market and glass-ceiling in the labor market.
An Iterative Approach to Estimation with Multiple High-Dimensional Fixed Effects, with Wenjia Zhu and Randall P. Ellis. August 21, 2017. Supplementary SAS program: Luo, Siyi, Wenjia Zhu, and Randall P. Ellis (2016), TSLSFECLUS.SAS ReadMe.pdf
Abstract
We develop a novel estimation algorithm for models with multiple high-dimensional fixed effects and unbalanced panels. Our algorithm absorbs fixed effects iteratively until they are asymptotically eliminated. Monte Carlo simulations show that our approach matches results from estimation with fixed effect dummies. Applying the algorithm to US employer-based health insurance data, we analyze health care utilization of 63 million individual-months with fixed effects for 1.4 million individuals, 150,000 primary care physicians, 3,000 counties, 465 employer*year*single/family coverage types and 47 months. We find that narrow network plans reduce the probabilities of monthly visits relative to preferred provider organizations.
Work in Progress
“Effects of Internal Migration on Income Distributions in China: Quantile Regression with Endogeneity”
Publication
Quantitative Modeling of Operational Risk, with Leyla Korkut, Mengxue Wang, Raymond T. Perkins III and Vincent Hong Chen. Risk Management, December 2013, Issue 28: 27-31. Published online by Society of Actuaries, Joint Risk Management Section.
