...
首页> 外文期刊>Lifetime Data Analysis >An extended cure model and model selection
【24h】

An extended cure model and model selection

机译:扩展的固化模型和模型选择

获取原文
获取原文并翻译 | 示例
   

获取外文期刊封面封底 >>

       

摘要

We propose a novel interpretation for a recently proposed Box-Cox trans formation cure model, which leads to a natural extension of the cure model. Based on the extended model, we consider an important issue of model selection between the mixture cure model and the bounded cumulative hazard cure model via the like lihood ratio test, score test and Akaike's Information Criterion (AIC). Our empirical study shows that AIC is informative and both the score test and the likelihood ratio test have adequate power to differentiate between the mixture cure model and the bounded cumulative hazard cure model when the sample size is large. We apply the tests and AIC methods to leukemia and colon cancer data to examine the appropriateness of the cure models considered for them in the literature.
机译:我们对最近提出的Box-Cox转化形成固化模型提出了一种新颖的解释,这导致了固化模型的自然扩展。在扩展模型的基础上,我们通过似然比测试,得分测试和Akaike的信息准则(AIC)考虑了在混合固化模型和有界累积危害固化模型之间进行模型选择的重要问题。我们的经验研究表明,AIC具有参考价值,当样本量较大时,得分测试和似然比测试都具有足够的能力来区分混合治愈模型和有界累积危害治愈模型。我们将测试和AIC方法应用于白血病和结肠癌数据,以检查文献中考虑的治愈模型的适当性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号