时间：2016年11月25日周五下午2:00
地点：统计楼103
报告人：王学钦（中山大学 教授）
题目：SURE INDEPENDENCE SCREENING ADJUSTED FOR CONFOUNDING COVARIATES WITH ULTRAHIGH DIMENSIONAL DATA
摘要：Detecting candidate genetic variants in genomic studies often encounters the confounding problems, particularly when the data are ultrahigh dimensional. Confounding covariates, such as age and gender, not only can reduce the statistical power, but also will introduce spurious genetic association. How to control for the confounders in ultrahigh dimensional data analysis is critical yet challenging issues. In this paper, we propose a novel sure independence screening method based on conditional distance correlation under the ultrahigh dimensional model setting. Our proposal accomplishes the adjustment by conditioning on the confounding variables. With the model-free feature of conditional distance correlation, our method does not need any parametric modeling assumptions and is thus quite flexible in practice. In addition, it is naturally applicable to data with multivariate response. We show that under some mild technical conditions, the proposed method enjoys the sure screening property even when the dimensionality is an exponential order of the sample size. The simulation studies and a real data analysis demonstrate that the proposed procedure has competitive performance.
个人简介：王学钦，中山大学数学学院和中山医学院双聘教授，博士生导师，中山大学统计学科带头人，中山大学华南统计科学研究中心执行主任，国家优秀青年基金获得者，教育部新世纪人才，教育部统计专业教指委委员。研究领域为非参多元统计学、统计学习、和精准医学。
报告人：林金官（南京审计大学 教授）
题目：A Robust and Efficient Estimation Method For Nonparametric Models with Mixed Discrete and Continuous Data
摘要：Nonparametric models with with mixed discrete and continuous regressors have applications to many fields, such as medicine, economics and finance. However, most ex-isting methods based on least squares or likelihood are sensitive due to the preference of outliers or the error distribution is heavy tailed. In this paper, a new robust and efficient estimation procedure based on local modal regression is proposed for nonparametric models with mixed discrete and continuous regressors, and a modified EM algorithm is introduced to estimate the proposed estimator. Under some regular conditions, the large sample theory is established for the proposed estimators. We show that the proposed estimators are as asymptotically efficient as the least-square based estimators when there are no outliers and the error distribution is normal. The simulations and real data analysis are conducted to illustrate the finite sample performance of the proposed method.