[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: str2ph and cluster
Run the program with and without using this option previously
to see whether it makes any difference.
On my understanding, "valid" is not necessarily the
best word here. -str2ph- is just a way of putting
a number on agreement between observed and predicted,
using a likelihood approach. That can be useful,
but it is not totemic.
> I have a question about Patrick Royston's str2ph command to
> calculate explained variation in survival analysis, which I am using
> with stcox.
> Are the results of str2ph still valid if the "cluster" option is
> used to adjust standard errors for intragroup correlation?
* For searches and help try:
- st: Totems
- From: "Campbell, Richard T." <email@example.com>