应数学与统计学院概率统计研究所邀请，四位国际统计学专家将于近期访问我校，期间将做系列学术讲座，具体安排如下（地点：齐云楼911多媒体报告厅）：
一、报告人：Zehua Chen （新加坡国立大学）
题目：Simultaneous Feature Selection and Precision Matrix Estimation in High-dimensional Multivariate Regression Models
时间：2015年7月31日上午9：00
摘要：We consider multivariate regression models with a q-dimensional response vector, a p-dimensional feature space and a sample of size n. We deal with the problem of feature selection and precision matrix estimation of the models in the case that both q and p are large compared with the sample size n. In theoretical consideration, we allow them to diverge to infinity as n goes to infinity. We give a conditional formulation of the multivariate regression model and propose an iterated alternate method which alternates at each iteration between a feature selection step and a precision matrix estimation step. At the feature selection step, we use a sequential feature selection procedure called sequential Lasso (SLasso). At the precision matrix estimation step, we adopt the neighborhood detection approach and use a sequential scaled pairwise selection (SSPS) method. We will discuss the detailed algorithm of the iterated alternate method as well as its asymptotic properties. Simulation studies comparing the iterated alternate method with other available methods will be presented. An application to a real data set will be reported as well.
二、报告人：杨广仁（暨南大学）
题目：Feature Screening in Ultrahigh Dimensional Cox's Model
时间：2015年7月31日上午11:00
摘要：Survival data with ultrahigh dimensional covariates such as genetic markers have been collected in medical studies and other fields. In this work, we propose a feature screening procedure for the Cox model with ultrahigh dimensional covariates. The proposed procedure is distinguished from the existing sure independence screening (SIS) procedures (Fan, Feng and Wu, 2010, Zhao and Li, 2012) in that the proposed procedure is based on joint likelihood of potential active predictors, and therefore is not a marginal screening procedure. The proposed procedure can effectively identify active predictors that are jointly dependent but marginally independent of the response without performing an iterative procedure. We develop a computationally effectivealgorithm to carry out the proposed procedure and establish the ascent property of the proposed algorithm. We further prove that the proposed procedure possesses the sure screening property. That is, with the probability tending to one, the selected variable set includes the actual active predictors. We conduct Monte Carlo simulation to evaluate the finite sample performance of theproposed procedure and further compare the proposed procedure and existing SIS procedures. The proposed methodology is also demonstrated through an empirical analysis of a real data example.
三、报告人：周望（新加坡国立大学）
题目：Estimate canonical correlation coefficients of high-dimensional normal vectors: finite rank case
时间：2015年7月31日下午2：30
四、报告人：刘志（澳门大学）
题目： Inference on the "pooled" high frequency data
时间：2015年7月31日下午4：30
摘要：In estimating the integrated volatility using high-frequency data, it is well documented that the presence of the micro-structure noise causes big challenge. In this paper, as demonstrated by the motivational simulation study, the common feature of multiple observations brings an additional problem to the estimation of the integrated volatility. It becomes one more source of bias in addition to the micro-structure noise. In this paper, we propose a multiplicity-adjusted and noise-corrected pre-averaging estimator which is proved to be consistent and have asymptotic normal distribution. Our approach is also easily extended to the case when the latent process has jumps. Extensive comparisons with empirical procedures in dealing with micro-structure noise and/or multiple transactions show that our newly proposed estimator is superior over others. Yet surprisingly, in some cases, our estimator performs even better than the ideal estimator which assumes the transaction times within a single time stamp are observable. Simulation studies justify our theory and we also implement our estimator to some real data sets.
欢迎届时光临！
应用数学与复杂系统重点实验室
兰州大学数学与统计学院
二〇一五年七月三十日
学校主页
