Variable Selection Procedures in Competing-Risks Models Using Penalized Likelihood

다국어 초록 (Multilingual Abstract)

Competing risks data commonly arise when an occurrence of an event precludes other type of events from being observed. Recently, competing-risks models with frailty have been studied for clustered competing-risks data that may be correlated. In this paper, we propose a variable selection procedure for fixed effects in the cause-specific hazard frailty model for the clustered competing-risks data using a penalized likelihood. Here we consider two popular penalty functions, least absolute shrinkage and selection operator (LASSO) and smoothly clipped absolute deviation (SCAD). We derive simple matrix forms for the variable selection procedure. The usefulness of the proposed method is illustrated using a practical example data set.

참고문헌 (Reference)

1 Fan J, "Variable selection via nonconcave penalized likelihood and its oracle properties" 96 : 1348-1360, 2001

2 Ha ID, "Variable selection in subdistribution hazard frailty models with competing risks data" 33 : 4590-4604, 2014

3 Ha ID, "Variable selection in general frailty models using penalized h-likelihood" 23 : 1044-1060, 2014

4 Wang H, "Tuning parameter selectors for the smoothly clipped absolute deviation method" 94 : 553-568, 2007

5 Prentice R, "The analysis of failure times in the presence of competing risks" 34 : 541-554, 1978

6 Tibshirani R, "Regression shrinkage and selection via the Lasso" 58 : 267-288, 1996

7 Sylvester R, "Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials" 49 : 466-477, 2006

8 Ha ID, "Model selection for multi-component frailty models" 26 : 4790-4807, 2007

9 Yuan M, "Model selection and estimation in regression with grouped variables" 68 : 49-67, 2006

10 Christian NJ, "Hierarchical likelihood inference on clustered competing risks data" 35 : 251-267, 2016

11 Ha ID, "Hierarchical likelihood approach for frailty models" 88 : 233-243, 2001

12 Lee Y, "Hierarchical generalized linear models (with discussion)" 58 : 619-678, 1996

13 Breiman L, "Heuristics of instability and stabilization in model selection" 24 : 2350-2383, 1996

14 Lee Y, "Generalised linear models with random effects: unified analysis via h-likelihood" Chapman and Hall 2006

15 Ha ID, "Estimating frailty models via Poisson hierarchical generalized linear models" 12 : 663-681, 2003

16 Ha ID, "Analysis of clustered competing risks data using subdistribution hazards models with multivariate frailties" 25 : 2488-2505, 2016

17 Huang J, "A selective review of group selection in high-dimensional models" 27 : 48199-, 2012

18 Fan J, "A selective overview of variable selection in high dimensional feature space" 20 : 101-148, 2010

19 Fine JP, "A proportional hazards model for the subdistribution of a competing risk" 94 : 496-509, 1999

20 Lee Y, "A new sparse variable selection via random-effect model" 125 : 89-99, 2014