Hello!
I run a clog-log regression and then look at the hazard function. Technically, that is what I do in Stata:
cloglog y i.x1#c.age X
qui margins i.x1 , at((asobserved) _all age=(16(1)45)) noestimcheck
marginsplot, xdimension(age)
In my understanding, predicting margins gives hazard estimates. It is very puzzling to me that there are negative values of hazard since cloglog supposably should have bouned them between 0 and 1.
What could be the explanation for this phenomenon?
Thank you in advance for your time and effort
I run a clog-log regression and then look at the hazard function. Technically, that is what I do in Stata:
cloglog y i.x1#c.age X
qui margins i.x1 , at((asobserved) _all age=(16(1)45)) noestimcheck
marginsplot, xdimension(age)
In my understanding, predicting margins gives hazard estimates. It is very puzzling to me that there are negative values of hazard since cloglog supposably should have bouned them between 0 and 1.
What could be the explanation for this phenomenon?
Thank you in advance for your time and effort
