Sample Size Calculation For Generalized Estimating Equations

The calculation of sample size based on generalized estimation equation has been widely used in various fields,and has also attracted the attention of researchers in various circles.For the data calculated based on the sample size of the generalized estimation equation,we study the longitudinal data with correlation among the data in the response variable cluster.In this paper,the calculation formula of sample size based on generalized estimation equation is derived,and the influence of different data structures and job-related structures on the fitting effect of sample size is explored,and the numerical simulation analysis is carried out.Finally,the actual data are analyzed.The research work of this paper mainly includes:(1)At the beginning,this paper introduces the background and significance of the topic selection,as well as the research status of the sample size calculation based on generalized estimation equation at home and abroad.(2)Logistic regression and generalized linear model were reviewed.Then,based on the particularity of the correlation in the research data cluster,the basic knowledge of the correlation of generalized estimated equation was introduced.(3)This paper introduces the calculation method and steps of sample size based on generalized estimation equation,and then deduces the calculation formula of sample size under three special data cases.(4)Simulation study was carried out based on the calculation formula of sample size derived in the case of binary response variables and horizontal covariables of clusters,and the influence of the number of clusters,the number of subunits and the misdesignation of work-related structure matrix on the estimated sample size was respectively studied and analyzed.(5)Case analysis of data from a clinical trial in patients with respiratory disease.