\(L_{1}\) penalized estimation in the Cox proportional hazards model. (English) Zbl 1207.62185

Goeman, Jelle J.

Biom. J. 52, No. 1, 70-84 (2010).

Summary: This article presents a novel algorithm that efficiently computes \(L_{1}\) penalized (lasso) estimates of parameters in high-dimensional models. The lasso has the property that it simultaneously performs variable selection and shrinkage, which makes it very useful for finding interpretable prediction rules in high-dimensional data. The new algorithm is based on a combination of gradient ascent optimization with the Newton-Raphson algorithm. It is described for a general likelihood function and can be applied in generalized linear models and other models with an \(L_{1}\) penalty. The algorithm is demonstrated in the Cox proportional hazards model, predicting survival of breast cancer patients using gene expression data, and its performance is compared with competing approaches. An R package, penalized, that implements the method, is available on CRAN.

MSC:

62N02	Estimation in survival analysis and censored data
62P10	Applications of statistics to biology and medical sciences; meta analysis
65K10	Numerical optimization and variational techniques
62J12	Generalized linear models (logistic models)
92C50	Medical applications (general)
65C60	Computational problems in statistics (MSC2010)

Keywords:

gradient ascent; lasso; penalty; survival

Software:

SparseLOGREG; penalized; LASSO; R

References:

[1]	Beer, Gene-expression profiles predict survival of patients with lung adenocarcinoma, Nature Medicine 8 pp 816– (2002)
[2]	BÃ\c{}velstad, Predicting survival from microarray data â a comparative study, Bioinformatics 23 pp 2080– (2007)
[3]	De Boer, Statistical analysis of sediment toxicity by additive monotone regression splines, Ecotoxicology 11 pp 435– (2002)
[4]	Efron, Least angle regression, Annals of Statistics 32 pp 407– (2004) · Zbl 1091.62054
[5]	Genkin, Large-scale Bayesian logistic regression for text categorization, Technometrics 49 pp 291– (2007)
[6]	Goeman, Testing association of a pathway with survival using gene expression data, Bioinformatics 21 pp 1950– (2005)
[7]	Gui, Penalized cox regression analysis in the high-dimensional and low-sample size settings, with applications to microarray gene expression data, Bioinformatics 21 pp 3001– (2005)
[8]	Keerthi, A fast tracking algorithm for generalized LARS/LASSO, IEEE Transactions on Neural Networks 18 pp 1826– (2007)
[9]	Kim, Y. and Kim, J., (2004). Gradient lasso for feature selection. In Proceedings of the 21st International Conference on Machine Learning, Volume 69 of ACM International Conference Proceeding Series, pp. 473â480.
[10]	Meier, The group lasso for logistic regression, Journal of The Royal Statistical Society Series B 70 pp 53– (2008) · Zbl 1400.62276 · doi:10.1111/j.1467-9868.2007.00627.x
[11]	Osborne, On the lasso and its dual, Journal of Computational and Graphical Statistics 9 pp 319– (2000)
[12]	Park, L1 regularization path algorithm for generalized linear models, Journal of the Royal Statistical Society Series B 19 pp 659– (2007) · doi:10.1111/j.1467-9868.2007.00607.x
[13]	Rosenwald, The use of molecular profiling to predict survival after chemotherapy for diffuse large-b-cell lymphoma, New England Journal of Medicine 346 pp 1937– (2002)
[14]	Segal, Microarray gene expression data with linked survival phenotypes: diffuse large-b-cell lymphoma revisited, Biostatistics 7 pp 268– (2006) · Zbl 1169.62388
[15]	Shevade, A simple and efficient algorithm for gene selection using sparse logistic regression, Bioinformatics 19 pp 2246– (2003)
[16]	Tibshirani, Regression shrinkage and selection via the LASSO, Journal of the Royal Statistical Society Series B 58 pp 267– (1996) · Zbl 0850.62538
[17]	Tibshirani, The LASSO method for variable selection in the Cox model, Statistics in Medicine 16 pp 385– (1997)
[18]	Van de Vijver, A gene-expression signature as a predictor of survival in breast cancer, New England Journal of Medicine 347 pp 1999– (2002)
[19]	Van Houwelingen, Cross-validated Cox regression on microarray gene expression data, Statistics in Medicine 25 pp 3201– (2006)
[20]	Van’t Veer, Gene expression profiling predicts clinical outcome of breast cancer, Nature 415 pp 530– (2002)
[21]	van Wieringen, Survival prediction using gene expression data: A review and comparison, Computational Statistics & Data Analysis 53 pp 1590– (2009) · Zbl 1453.62225
[22]	Verweij, Cross-validation in survival analysis, Statistics in Medicine 12 pp 2305– (1993)
[23]	Wang, Gene-expression pro-files to predict distant metastasis of lymph-node-negative primary breast cancer, Lancet 365 pp 671– (2005) · doi:10.1016/S0140-6736(05)17947-1
[24]	Yuan, Model selection and estimation in regression with grouped variables, Journal of The Royal Statistical Society Series B 68 pp 49– (2006) · Zbl 1141.62030
[25]	Zou, Regularization and variable selection via the elastic net, Journal of the Royal Statistical Society Series B 67 pp 301– (2005) · Zbl 1069.62054
[26]	Zou, On the âdegrees of freedomâ of the lasso, Annals of Statistics 35 pp 2173– (2007)