A Composite Likelihood Method for the Analysis of Multivariate Survival Data: An Application to a PBDE Study

When dealing with multivariate survival data, featuring the association structure is a key difference from the univariate survival analysis. In this paper, we explore to use the composite likelihood framework to handle multivariate survival data, where only the lower dimensional survival distributions need to be specified. The development allows us to use available modeling schemes for bivariate survival data to characterize association structures of correlated survival times. The inference procedure is based on the pseudolikelihood which is the product of the lower dimensional bivariate distributions. The proposed estimation procedure is assessed through simulation studies. As a genuine application, we apply the composite likelihood inference procedure to analyze the data from the polybrominated diphenyl ethers (PBDEs) study, where four types of PBDE congeners are available. The associations among the four PBDE congeners, and the relationships between the covariates and the PBDE congeners are of interest. The result shows that there is strong association among the concentrations of the four PBDE congeners, and statistically significant predictors on the concentrations of the four PBDE congeners are identified.

LOESS smoothed density estimates for multivariate survival data subject to censoring and masking

Actuaries often encounter censored and masked survival data when constructing multiple-decrement tables. In this paper, we propose estimators for the cause-specific failure time density using LOESS smoothing techniques that are employed in the presence of left-censored data, while still allowing for right-censored and exact observations, as well as masked causes of failure. The smoothing mechanism is incorporated as part of an expectation-maximisation algorithm. The proposed models are applied to a bivariate African sleeping sickness data set.