Lifetime data analysis is a major area of research in statistics. Multi-state models are commonly used to characterize life history processes. In this session speakers will discuss some important research questions under different types of multi-state models for the analysis of lifetime data. Dr. Paul Peng from Queen’s University is going to present an alternative modeling approach to the commonly used proportional hazards frailty models to analyze clustered and recurrent failure times data. Dr. Yongzhao Shao from New York University is going to discuss a regularized competing risk regression to assess heterogeneous effects on competing outcomes in mortality analysis. Dr. Leilei Zeng from University of Waterloo is going to talk about a mixture hidden Markov model for multiple types of disease. All speakers have agreed to participate. Summaries of their talks are as follows:
1. Speaker: Dr. Yingwei (Paul) Peng, Professor in Department of Public Health Sciences, Queen’s University, Kingston, ON, Canada
Title: Additive hazards frailty models with semi-varying coefficients
Summary: Proportional hazards frailty models have been extensively investigated and used to analyze clustered and recurrent failure times data. However, the proportional hazards assumption in the models may not always hold in practice. In this paper, we propose an additive hazards frailty model with semi-varying coefficients, which allows some covariate effects to be time-invariant while other covariate effects to be time-varying. The time-varying and time-invariant regression coefficients are estimated by a set of estimating equations, whereas the frailty parameter is estimated by the moment method. The large sample properties of the proposed estimators are established. The finite sample performance of the estimators is examined by simulation studies. The proposed model and estimation are illustrated with an analysis of data from a rehospitalization study of colorectal cancer patients.
2. Speaker: Dr. Yongzhao Shao, Professor in Division of Biostatistics, Department of Population Health, NYU Grossman School of Medicine, New York, NY, USA
Title: Assessing heterogeneous effects on competing outcomes in mortality analysis
Summary: There is often a need to evaluate heterogeneous effects on competing survival events due to different causes, increasingly so in competing-risk survival analysis of late-onset diseases. In such problems, it is not only of interest to identify the predictors that are significantly associated with the competing outcomes, but also to identify any predictors that have different effects among the competing survival events. We propose a regularized competing risk regression to achieve the heterogeneity pursuit and model selection simultaneously. We develop a computing algorithm and show that our approach can achieve estimation consistency. Simulation studies further demonstrate the effectiveness of our method under various practical scenarios. We will discuss applications to mortality analysis in Alzheimer's disease research to understand the heterogeneous effects the allele 4 of the ApoE gene in the context of competing risk analysis.
3. Speaker: Dr. Leilei Zeng, Associate Professor in Department of Statistics and Actuarial Science, University of Waterloo, Waterloo, ON, Canada
Title: A mixture hidden Markov model for multiple types of disease
Brief Summary: This study concerns disease subtype heterogeneity, and we propose a hidden multistate model with an underlying process being a finite mixture of Markov processes, one for each disease subtype.