Statistical analysis of high dimensional gene data

Abstract

We consider the problem of constructing an additive risk model based on the right censored survival data with high dimensional covariates to predict the survival times of the cancer patients. We apply Partial Least Squares to reduce the dimension of the covariates and get the latent variables, say, components; these components can be used as new regressors to fit the extensional additive risk model. Also the time dependent AUC curve (area under the receiver operating characteristic (ROC) curve) is employed to assess how well the model predicts the survival time. This approach is illustrated by analysis of breast cancer dataset.