Scholarlink

<p>In this paper, we consider ridge-type shrinkage estimators to estimate the coefficients in the low and high-dimensional Cox proportional hazards regression model in the presence of multicollinearity. The proposed estimators are evaluated analytically using their asymptotic bias and risk. Moreover, we consider two penalized estimators, namely, the LASSO and Elastic-Net and compare their relative performance with the other proposed estimators numerically. Monte-Carlo simulation experiment and real data analysis are conducted to evaluate the performance of each estimator based on the simulated relative efficiency. The results show that both in low and high-dimensional data, new shrinkage estimators dominate the usual ridge estimator that discards weak predictors. We also find that developed methods would be useful for the practitioners in various sciences.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Efficient and improved ridge-type shrinkage estimators in low and high dimensional cox proportional hazards regression model

  • 期刊论文

  • 发表日期：

    2025年5月19日

查看全文
  • Seyedeh Zahra Aghamohammadi

摘要

In this paper, we consider ridge-type shrinkage estimators to estimate the coefficients in the low and high-dimensional Cox proportional hazards regression model in the presence of multicollinearity. The proposed estimators are evaluated analytically using their asymptotic bias and risk. Moreover, we consider two penalized estimators, namely, the LASSO and Elastic-Net and compare their relative performance with the other proposed estimators numerically. Monte-Carlo simulation experiment and real data analysis are conducted to evaluate the performance of each estimator based on the simulated relative efficiency. The results show that both in low and high-dimensional data, new shrinkage estimators dominate the usual ridge estimator that discards weak predictors. We also find that developed methods would be useful for the practitioners in various sciences.