PLOTS AND TESTS FOR GOODNESS OF FIT WITH RANDOMLY CENSORED-DATA
成果类型：
Article
署名作者：
NAIR, VN
刊物名称：
BIOMETRIKA
ISSN/ISSBN：
0006-3444
DOI：
10.2307/2335810
发表日期：
1981
页码：
99103
关键词：
摘要：
Nonparametric procedures are considered for assessing goodness of fit when the data may be subject to random censoring. Two classes of large-sample tests are introduced, based, respectively, on the maximum and average weighted difference between the specified survival function and the Kaplan-Meier estimate of the true survival function. The procedures in the 1st class are inverted to obtain simultaneous confidence bands for the survival and cumulative hazard functions. These bands provide simultaneous bands associated with the percentage (P-P), quantile (Q-Q) and hazard plots for censored data.