| 主讲人 |
Peter Song, University of Michigan |
简介 |
<p><span style="font-size: small;"><span style="font-family: Arial;"> Abstract:</span></span></p>
<p><span style="font-size: small;"><span style="font-family: Arial;"><span lang="EN-US">This article is to discuss the role of generalize method of moments (GMM) in parameter estimation and statistical inference along with the strategy of divide-and-combine for Big Data analysis. As an effective inferential tool, Efron's confidence distribution (CD) has attracted a surge of renewed attention. The essence in constructing confidence distribution pertains to the availability of suitable pivotal quantities, which are usually obtained from the (asymptotic) distribution of point maximum likelihood estimator. We propose to use inference function, from which the parameter is obtained, as the basis of constructing the pivotal. The proposed method, termed as merged estimating function analytics (MEFA), inherits several advantages of inference function over the traditional likelihood reduced score function. We show that the proposed MEFA is a special case of the generalized method of moments (GMM). Thus, MEFA, which includes maximum likelihood estimation as a special case, provides us a unified framework for many kinds of statistic methods, which is illustrated via numerical examples in the context of divide-and-combine approaches to Big Data analysis. </span></span></span></p> |
