Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/79573 
Year of Publication: 
2012
Series/Report no.: 
SFB 649 Discussion Paper No. 2012-061
Publisher: 
Humboldt University of Berlin, Collaborative Research Center 649 - Economic Risk, Berlin
Abstract: 
We deal with two kinds of Cox regression models with varying coefficients. The coefficients vary with time in one model. In the other model, there is an important random variable called an index variable and the coefficients vary with the variable. In both models, we have p-dimensional covariates and p increases moderately. However, it is the case that only a small part of the covariates are relevant in these situations. We carry out variable selection and estimation of the coefficient functions by using the group SCAD-type estimator and the adaptive group Lasso estimator. We examine the theoretical properties of the estimators, especially the L2 convergence rate, the sparsity, and the oracle property. Simulation studies and a real data analysis show the performance of these new techniques.
Subjects: 
Cox regression model
high-dimensional data
sparsity
oracle estimator
B-splines
group SCAD
adaptive group Lasso
L2 convergence rate
JEL: 
C14
C24
Document Type: 
Working Paper

Files in This Item:
File
Size
440.85 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.