The Censored Regression Model Under Complex Survey And Its Application In Medical Study

Objective to study the method of variance estimation when we use complex sample to fit linear regression,censored regression model including COX proportional hazards regression model and tobit regression model, then deal with medical problems using these methods.Methods to simulate complex sample,then fit linear regression, COX proportional hazards regression model and tobit regression model with and without taking sampling characteristics into account respectively, compute the mean and standard error of regression coefficient and compare the differences of these two results, giving a COX regression example at the same time.Results with a fixed sample size,the regression coefficient of linear regression,COX proportional hazards regression model and tobit model taking into account of the sampling characteristics is more closer to the true value, but the standard error of regression coefficient is different from that of ignoring the sampling characteristics. If the units in the cluster is highly correlated,the standard error will be larger than that of lower intra-cluster correlation coefficient are highly heterogeneity, the standard error of regression coefficient, but the variance of estimator taking into the sampling characteristics is lower than that of ignoring the sampling characteristics.Conclusion we should take into account the sampling characteristics when dealing with complex survey data, and estimate relevant variance using appropriate techniques so as to get more reliable results.