A0254
Title: BRBVS: variable ranking in copula survival models affected by general censoring scheme
Authors: Danilo Petti - University of Essex (United Kingdom) [presenting]
Marcella Niglio - University of Salerno (Italy)
Marialuisa Restaino - University of Salerno (Italy)
Abstract: The newly developed BRBVS package is introduced and discussed. BRBVS presents an innovative approach to variable selection in the presence of bivariate time-to-event data, characterized by censoring/truncation and correlation. This tool allows researchers to identify two sets of relevant covariates by a new metric that considers the dependency structure between survival functions. The effectiveness of BRBVS is demonstrated through numerical and graphical results from an extensive simulation study and with the analysis of a data set collected from a study on age-related eye disease.
