4 fix wrong description of what i do
source | link

One way would be not to takegenerate the mean ofcross-validated survival after each repeat, but instead just gather all the 50 (cross-validated) risks for each patient and take the mean of it. With these means i could create two new survival curves and take their loglog-rank statistics. By this i get a bunch of p-Values and as a first step I just took the median of these values, to get some way a "median difference"Which sounds somehow reasonable. I am sure this is not really valid, but i wonder, what would be a correct way to compare the survival curves and get some "p-Value"?

One way would be to take the mean of all 50 (cross-validated) risks for each patient. With these i could create new survival curves and take their log-rank statistics. By this i get a bunch of p-Values and as a first step I just took the median of these values, to get some way a "median difference". I am sure this is not really valid, but i wonder, what would be a correct way to compare the survival curves and get some "p-Value"?

One way would be not to generate the cross-validated survival after each repeat, but instead just gather all the 50 (cross-validated) risks for each patient and take the mean of it. With these means i could create two new survival curves and take their log-rank statistics. Which sounds somehow reasonable.

3 added 199 characters in body
source | link

One way would be to take the mean of all 50 (cross-validated) risks for each patient. With these i could create new survival curves and take their log-rank statistics.

Would By this i get a bunch of p-Values and as a first step I just took the median of these values, to get some way a "median difference". I am sure this is not really valid, but i wonder, what would be a correct way to compare the survival curves and get some "p-Value"?

One way would be to take the mean of all 50 (cross-validated) risks for each patient. With these i could create new survival curves and take their log-rank statistics.

Would this be a correct way to compare the survival curves?

One way would be to take the mean of all 50 (cross-validated) risks for each patient. With these i could create new survival curves and take their log-rank statistics. By this i get a bunch of p-Values and as a first step I just took the median of these values, to get some way a "median difference". I am sure this is not really valid, but i wonder, what would be a correct way to compare the survival curves and get some "p-Value"?

2 added 184 characters in body
source | link

I have around 50 patients and their survival times and their (say clinical) measurements. The goal is to use the measurements to find a model that can predict whether a new patient will have low or high risk "to survive". Because of the low number of patients, the goal is just to make sure that the (cross-validated, not single) model can "significantly" distinguish between the low and the high risk group.

I have around 50 patients and their survival times and their (say clinical) measurements. The goal is to use the measurements to find a model that can predict whether a new patient will have low or high risk "to survive".

I have around 50 patients and their survival times and their (say clinical) measurements. The goal is to use the measurements to find a model that can predict whether a new patient will have low or high risk "to survive". Because of the low number of patients, the goal is just to make sure that the (cross-validated, not single) model can "significantly" distinguish between the low and the high risk group.

1
source | link