Subgroup Analysis Of Medical Longitudinal Data

Longitudinal data refers to the data which acquired by a series of test individuals following up with time.It often occurs in disciplines,such as psychology,sociology,medicine and biology.The remarkable feature of longitudinal data is that it combines the cross-sectional data and time series data.It can not only analyze the trend of observation units changing with time,but also analyze the general trend of change.In recent years,the study of longitudinal data has aroused wide attention between domestic scholars and foreign scholar.In this paper,two heterogeneity models longitudinal data are established,and the alternating direction method of multipliers algorithm is adopted to achieve the dual purpose of identifying subgroups and parameter estimation.Medically,personalized medical treatment has become a hot spot in new medical research in recent years,because of the higher cure rate.The key step in personalized medicine is to be able to identify the subgroups of the mixed population so that each subgroup can be treated accurately.Understanding the importance of the heterogeneity of treatment for precision medical development,and precision medical treatment can also be used to find suitable medical treatments for patients with similar characteristics.A major challenge is that we usually do not have a priori knowledge of the information of the patient group.In order to solve this problem,we first assume that the variable coefficient in treatment is dependent on the subgroup of patients,and belongs to a different subset of information are unknown,consider a mixed effect regression model;then,under the assumption that the sources of heterogeneity in potential unobserved variables under the condition that the subject specific intercept the potential factors of heterogeneity,a fixed effect model.In order to estimate the model parameters and the identification of subgroups,the paper,using an automatic alternating direction estimation group structure and subgroup specificmultiplier method,this algorithm under certain conditions,the real group information a priori knowledge of the ORACLE least squares estimator is the local minimum of the objective function in high probability under that provides a theoretical support to statistical inference of this nature as a subgroup number and model parameters.Finally,we use stochastic simulation to estimate the loss parameters,and illustrate the practical effect of subgroup analysis by analyzing data from a group of AIDS group published by Kalsow.