您所在的位置: 首页» 新闻中心» 讲座预告

11-21 Bayes risk of mean excess: Theory and applications

报告题目:Bayes risk of mean excess: Theory and applications

主讲人:Ehsan S. Soofi




        EHSAN S. SOOFI is a distinguished professor of Management Science and Statistics at Sheldon B. Lubar School of Business, University of Wisconsin-Milwaukee. He got his BA in mathematics (1969), MA in statistics (1974) and PhD in applied statistics (1985) from University of California (Los Angeles, Berkeley, Riverside). He is a Fellow of the American Statistical Association, a Fellow of the Info-Metrics Institute. His research interests are on Information-theoretic and Bayesian modeling, applications to reliability analysis, management science, and economics. His publications are appeared in the following journals:

        Journal of the American Statistical Association

        European Journal of Operational Research

        Journal of Applied Probability,

        Econometric Reviews

        Strategic Management Journal

        Reliability Engineering & System Safety

        Naval Research Logistics

        Statistical Science

        Probability in the Engineering and Information Sciences

        Computational Statistics and Data Analysis

        Journal of Econometrics

        And others.


        The Bayes risk of the mean residual (MR) or excess of a random variable above a threshold is defined as the expectation with respect to a prior for the threshold. We show that the standard deviation provides a tight upper bound for the Bayes risk of the MR. This result characterizes the exponential distribution and is tighter than a known bound for the measure. The ratio of Bayes risk of the MR to the standard deviation provides a risk index which reached the maximum of one if and only if the underlying distribution is exponential. Distributional examples illustrate the tightness of the standard deviation bound, improvement over the known bound, and the risk index. I present an overview of recent results on the Bayes risk of the MR and describe applications to reliability of system, ranking forecast models, and New York City's taxi trip times.


XML 地图 | Sitemap 地图