重複事件與截切時間獨立假設檢定

長期追蹤研究常會收集某時間內一人發生多次事件的重複事件資料，例如電腦一再損壞送修、車禍一再發生。當某些個體之事件發生於研究開始前，就會產生截切重複事件資料。大多數研究截切重複事件資料的統計方法都假設重複事件時間與截切時間在可觀察之區域內部分獨立 (quasi-independence)，然而很多情況下此假設不一定合理。本文我們針對截切重複事件且右設限之資料，延伸conditional Kendall’s tau之概念提出一無母數檢定統計量以檢定此部分獨立假設。模擬以呈現本檢定之表現。最後以實例說明本檢定之應用。

關鍵字

重複事件；截切；部分獨立檢定

並列摘要

Repeated events data are encountered in many longitudinal studies such as computer breakdowns and motorcycle crashes that can occur repeatedly.over time for each subject. Truncated repeated events data often arises when subjects experience a certain event prior to study. Most statistical methods for the truncated repeated events data depend on the assumption of quasi-independence that repeated event times and truncation time are independent in the observable region. However, in many applications, this assumption might be unrealistic. In this article, we propose a nonparametric test statistic to test the quasi-independence assumption for truncated repeated event data in the presence of right censoring, which is an extension of the conditional Kendall’s tau. Simulation studies are conducted to investigate the performance of the proposed test statistic. A real example is also presented to illustrate the proposed test.

並列關鍵字

repeated events ； truncation ； test for quasi-independence

參考文獻

1. Chen, C. H., Tsai, W. Y. ＆ Chao, W. H. (1996). The Product-moment Correlation Coefficient and Linear Regression for Truncated Data. J. Am. Statist. Assoc. 91, 435-450

2. Kendall, M. G. (1938). A new measure of rank correlation. Biometrika 30, 81-93

3. Lin, M. R., Chang, S. H., Pai, L. ＆ Keyl, P. M. (2003). A longitudinal study of risk factor for motorcycle crashes among junior college students in Taiwan. Accident Analysis＆Prevention 35, 243-252

4. Tsai, W. Y. (1990). Testing the assumption of independence of truncation time and failure time. Biometrika 77, 167-177

5. A. Salter, P. J. Solomon (1997). Truncated Recurrent Event Survival Models for Methadone Data. Biometrics 53, 1293-1303

國際替代計量

重複事件與截切時間獨立假設檢定

全文下載

主題瀏覽