Details:
Speaker: Yichen Qin, University of Cincinnati
Title: Maximum Tangent Likelihood Estimation and Robust Regression
Abstract: In this article, we introduce a new robust estimation procedure for linear regression called maximum tangent likelihood estimation. By robustifying the traditional log-likelihood function, we are able to generate a class of robust estimates of the regression coefficients by adjusting the trade-off between efficiency and robustness. Furthermore, we propose a penalized tangent likelihood estimation for variable selection. We prove the consistency and oracle properties for such a method and show that it achieves the breakdown point of 0.5. We demonstrate the superior performance of our estimator by several simulation studies as well as real data examples.
Faculty host: Guanqun Cao
Brief Description of the Speaker’s Academic and Professional Achievements/Credentials:
Dr. Yichen Qin received a Bachelor in Statistics at Renmin University in 2005 and a Master of Science in Statistics at Columbia University in 2007. After that, he studied with Dr. Carey E. Priebe, a Senior Member of the IEEE, a Lifetime Member of the Institute of Mathematical Statistics, and a Fellow of the American Statistical Association, in Johns Hopkins University. Qin obtained a Ph.D. in Applied Mathematics and Statistics from Johns Hopkins University in 2013. In the same year, he joined the Department of Operations, Business Analytics, and Information Systems in the Lindner College of Business at the University of Cincinnati as an Assistant Professor.
His current line of research focuses on computational statistics, robust statistics, clustering analysis, mixture models, variable selection, financial statistics, social network analysis. He has published over 8 papers in high impact statistics journals including Journal of the American Statistical Association, Statistics in Medicine, etc.
|
