Contal C, Mallet A
Unité INSERM U436, CHU Pitié-Salpêtrière, 91 Bd de l'Hôpital, 75634 Paris Cedex 13, France.
Rev Epidemiol Sante Publique. 2000 Oct;48(5):490-501.
The Cox model is based on the proportional hazards hypothesis. In real analyses, this hypothesis may not hold. To evaluate the relevance of an interpretation of data using the Cox model, it is important to check this assumption. The major goodness-of-fit methods are reviewed.
Cox模型基于比例风险假设。在实际分析中,这一假设可能不成立。为了评估使用Cox模型对数据进行解释的相关性,检验这一假设很重要。本文回顾了主要的拟合优度方法。