Hello all,

I have 8 pairs of observations (each pair comes from the same person). For each observation I have 5 time points, and in each time point the occurrence of the "event" is checked. Similar to log rank test, I want to compare the two survival curves. The problem is, I don't have 2 independent groups of 8 observations each, I have 8 observations in each treatment, but they come in pairs. I do not know the level of correlation now. Is there is a way to do it ?

Thank you.