Capturing simple and complex time dependent effects using flexible parametric survival models A simulation study.pdf (2.72 MB)
Capturing simple and complex time-dependent effects using flexible parametric survival models: A simulation study
journal contributionposted on 2019-07-18, 11:20 authored by H Bower, M Crowther, M Rutherford, T Andersson, M Clements, X-R Liu, D Paul, P Lambert
Non-proportional hazards are common within time-to-event data and can be modeled using restricted cubic splines in flexible parametric survival models. This simulation study assesses the ability of these models in capturing non-proportional hazards, and the ability of the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) in selecting degrees of freedom. The simulation results for scenarios with differing complexities showed little bias in the survival and hazard functions for simple scenarios; bias increased in complex scenarios when fewer degrees of freedom were modeled. Neither AIC nor BIC consistently performed better and both generally selected models with little bias
Michael J. Crowther was in part funded by a National Institute for Health Research (NIHR) Doctoral Research Fellowship [DRF-2012-05-409]. Therese M-L. Andersson was supported by the Swedish Research Council [521-2011-3205] and the Swedish E-Science Research Center. Paul C. Lambert was supported the Swedish Research Council [521-2013-3383] and the Swedish Cancer Society [CAN2012/75Y]. Mark Clements was supported by the Swedish Cancer Society [CAN2012/765]. Swedish Research Council (Vetenskapsrådet in Swedish), Swedish Cancer Society (Cancerfonden in Swedish).
CitationCommunications in Statistics - Simulation and Computation, 2019
Author affiliation/Organisation/COLLEGE OF LIFE SCIENCES/School of Medicine/Department of Health Sciences
- VoR (Version of Record)