| Summary: | 碩士 === 東海大學 === 統計學系 === 102 === In clustered survival data, subunits within each cluster share similar characteristics, so that observations made from them tend to be positively correlated. In clinical trials, the correlated subunits from the same cluster are often randomized to different treatment groups. In the case, for testing the equality of marginal survival distributions between two treatment groups, we need to consider the intracluster correlations both within and between treatment groups. Rank tests for the difference or ratio of median failure time for paired observations have been proposed. Therefore, in this paper, we consider the problem of constructing the conditional confidence intervals for the difference of two median survival times given the covariates under a shared gamma frailty model for clustered survival data. Based on the conditional confidence intervals, we can identify the possible range of covariates over which the two groups would provide different median survival times. The implementation of the confidence intervals will be illustrated using real data sets. The associated coverage probability and the length of the interval will be investigated via a simulation study.
|
|---|
