报告题目：A generalized Lindley-binomial model for analyzing the proportions with end-point inflation

报告时间：2023年6月28日 下午 15:00

报告地点：南湖校区教学科研楼104

主办单位：科研处/数学与统计学院

主讲人：邓殿良

邓殿良简介：加拿大里贾纳大学数学与统计系的终身教授。吉林大学数学学位和概率与数理统计专业硕士学位,温莎大学统计学硕士和博士学位，主要从事概率论的理论研究和统计学及其在生物和医学科学中的应用,在《英国皇家统计学会杂志(B)》、《多元分析杂志》、《医学统计》、《数学分析与应用杂志》、《非参数统计》、《医学研究统计方法》和《生物制药统计学杂志》等杂志上发表学术论文 50余篇。

摘要：In this paper, a new three-parameter probability distribution called

Generalized Lindley-binomial (GLB) distribution is proposed to analyze

the proportional data with end-point inflation. The probabilistic properties

of the distribution such as the probability mass function, the rth factorial

moment function are derived. The Fisher scoring algorithm and the EM

algorithm are presented for computing estimations of parameters in the GLB model with/without covariates. Further the methods on model diagnosis and goodness of fit for the proposed model are discussed. A limited simulation study is performed to demonstrate the performance of derived EM algorithms for estimating parameters in the GLB model with/without covariates. The whitefly dataset is used to present the application to the proposed Lindley binomial regression model.