Regularisation in discrete survival models : a comparison of lasso and gradient boosting (original) (raw)

Abstract

We present the results of a simulation study performed to compare the accuracy of a lassotype penalization method and gradient boosting in estimating the baseline hazard function and covariate parameters in discrete survival models. The mean square error results reveal that the lasso-type algorithm performs better in recovering the baseline hazard and covariate parameters. In particular, gradient boosting underestimates the sizes of the parameters and also has a high false positive rate. Similar results are obtained in an application to real-life data.

Loading...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

References (10)

  1. B , D., M , M., B , B., W , S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67, 1-48.
  2. B , P. J , J. (2015). Group descent algorithms for nonconvex penalized linear and logistic regression models with grouped predictors. Statistics and Computing, 25, 173-187.
  3. E , B. (1988). Logistic regression, survival analysis, and the Kaplan-Meier curve. Journal of the American Statistical Association, 83, 414-425.
  4. F , L. (1994). Dynamic modelling and penalized likelihood estimation for discrete time survival data. Biometrika, 81, 317-330.
  5. F , Y. S , R. E. (1996). Experiments with a new boosting algorithm. In Proceedings of the Thirteenth Conference on Machine Learning, Bari, Italy. 148-156.
  6. F , J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, 29, 1189-1232.
  7. F , J. H., H , T., T , R. (2000). Additive logistic regression: A statistical view of boosting. Annals of Statistics, 28, 337-407.
  8. F , J. H., H , T., T , R. (2010). Regularization paths for generalized linear models via coordinate descent. Journal of Statistical Software, 33, 1-22.
  9. G , J. J. (2010). L1 penalized estimation in the Cox proportional hazards model. Biometrical Journal, 52, 70-84.
  10. G , A. (2011). glmmLasso: Variable selection for generalized linear mixed models by L1- A COMPARISON OF LASSO AND GRADIENT BOOSTING