[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
I have two related questions regarding STATA's cumulative hazard calculations when it produces the chart generated by the command stcurv,cumhaz following a Weibull or Gamma parametric regression.
1) Do the two numbers correspond to the minimum and maximum cumulative hazards? In some instances I am getting a nonzero, and in others a zero number for the lower value on the chart.
2) If these are the min and max, why are they not the same as when I use the predict,hazard command and manually sum up the hazards when the covariates are set to their mean values? Perhaps STATA evaluates a continuous function (integral) in calculating its cumulative hazards, resulting in the discrete sum over the time interval being different?
I'd appreciate any help you may be able to provide.
* For searches and help try: