报告题目：A Lindley-binomial model for analyzing the proportions with sparseness and excessive zeros
报 告 人：邓殿良 教授 University of Regina
报告时间：2023年6月20日 15:00-16:00
报告地点：数学楼第二报告厅
校内联系人：程建华 chengjh@jlu.edu.cn
报告摘要：In this paper, we present a new modelling approach for the proportions with sparseness and excessive zeros. The distribution of proportional data typically exhibits overdispersion, zero-inflation and sparseness and heavy tails. We propose a new Lindley binomial distribution, by compounding the two-parameter Lindley family of distributions with the binomial distribution. This distribution can flexibly handle each of the aforementioned features of proportional data. We study the probabilistic properties of this distribution such as moment, moment generating function and develop a computational approach to accurately evaluate the likelihood of proposed model and to perform the penalized maximum likelihood estimation via EM algorithm. We assess the performance of our developed algorithm for the estimation of parameters in the proposed model with/without covariates and demonstrate the application to Incidence of Hepatitis A and Whitefly data.
报告人简介：Dr. Deng is a full Professor at the Department of Mathematics and Statistics, University of Regina. He got his Bachelor degree in Mathematics and Master degree in Probability and Mathematical Statistics both from Jilin University, China. He then continued to pursue his graduate studies in Canada and obtained his Master and PhD degrees in Statistics from University of Windsor.
Dr. Deng is mainly engaged in theoretical research of probability theory and statistics and its application in biological and medical sciences. He has published about 50 academic papers in peer reviewed journals including Journal of Royal Statistical Society (B), Journal of Multivariate Analysis, Statistics in Medicine, Journal of Mathematical Analysis and Applications, Statistical Methods in Medical Research and Journal of Biopharmaceutical Statistics. He is reviewer and/or referee for more than 20 journals such as J. Nonp. Stat., Can. J. Stat., Biometrics, Stat. Med., J. Multi. Anal., J. Appl. Stat., J. Stat Comput. Simul., Stat. Sinica, Ann. Appl. Stat., Stat. Methods in Med. Res., Appl. Math. Modelling, and Stat. and Its Interface.
