e**h
发帖数: 46 | 1 向版上大牛们请教个问题:
我有两组肿瘤病人，分别用两种治疗方法。用logrank test检测两组病人存活时间是否
相同。但是logrank test 只能得到two-sided pvalue。现在我们需要one-sided
pvalue。该怎么计算？请版上大牛推荐个方法和学习资料。包子感谢。 | t*******t
发帖数: 633 | 2 http://en.wikipedia.org/wiki/Logrank_test
According to WIKI, the distribution of the test statistic is asymptotic
normal, which means it's symmetric, so the one sided p-vlaue=two sided p-
value/2.
【在 e**h 的大作中提到】
: 向版上大牛们请教个问题:
: 我有两组肿瘤病人，分别用两种治疗方法。用logrank test检测两组病人存活时间是否
: 相同。但是logrank test 只能得到two-sided pvalue。现在我们需要one-sided
: pvalue。该怎么计算？请版上大牛推荐个方法和学习资料。包子感谢。
| e**h
发帖数: 46 | 3 haha, I got it. Thanks.
However, be careful, the logrank test is not normally distributed, but chis-
quare, under h0. Only the logrank statistic, +/- sqrt(logrank test), is
approximately standard normal.
I will send you baozi |
|
