Each week, we publish layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.
The article featured today is from the Canadian Journal of Statistics with the full article now available to read here.
Liu, P., Song, S. and Zhou, Y. (2022), Semiparametric additive frailty hazard model for clustered failure time data. Can J Statistics. https://doi.org/10.1002/cjs.11647
In this paper, the authors propose a new model to modelling the clustered failure time data while there is intra-cluster dependence among the same cluster. The clustered failure time data is frequently encountered in biomedical and clinical studies recently. For example, in the western Kenya parasitemia study, where the researcher is interested in the time from a child’s enrolment to the infection of parasitemia, 542 children from 309 households were enrolled, the children from the same household forms a cluster, the failure time among the same cluster were extremely likely to be correlated since they share the similar gene and expose to the same environment. In the study, the covariates included baseline parasitemia density (BPD) and exposure to mosquito bites (BITE), Age and Gender, and the purpose of the model is to model how these covariates will affect failure time, while still take the intra-cluster dependence. And the response is hazard function instead of failure time. We model the relationship under the additive structure and assume the intra-cluster dependence acts additively on the hazard function. Some of the covariate effect is time-varying (such as BPD) while some are not (BITE, Age and Gender). The authors proposed a simple method to estimate the parameters and showed that the convergence rate of the constant covariates can be improved by integration technique. The estimator has a closed-form and can easily be implemented in practice. They also studied the large sample properties and finite sample performance. They applied their method to the western Kenya parasitemia study as well.