Semiparametric Approach to Quantile Regression for Random Coefficients Models

• Activity Number: 619
• Type: Topic Contributed
• Date/Time: Thursday, August 5, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Mi-Ok Kim*+
• Companies: Cincinnati Children's Hospital Medical Center
• Keywords: Random Coefficient Models ; Quantile Regression ; Empirical Likelihood ; Markov Chain Monte Carlo

Abstract

We consider quantile regression for random coefficients models with clustered data. Cluster specific random coefficients represent regression quantiles at cluster levels and their means represent population level average of conditional regression quantiles of interest. We take a semi-parametric approach to the estimation and related inference. The estimation of the conditional quantiles at the cluster level is formulated as an estimating equations problem and empirical likelihood is used to incorporate parametrically specified random effects in the estimating equations problem. We use Markov Chain Monte Carlo (MCMC) samplers for the computation.