TEST-Using stcrreg with cluster robust standard error and iweights - is this a valid approach?

[Mark Marshall]

Dear all,

OSX, Stata MP 14.1

stcox does not allow cluster robust standard error and iweights. I am not sure whether this is because it is a fundamentally unsound mathematical approach, or whether is simply computationally unfeasible.

stcrreg does allow this combination, if specified. Is this a valid approach, and should I specify it if allowed?

Thanks,

Mark