Strange computatonal problem with the Traj command

[Luis Mijares Castaneda]

Hello, I am trying to use the dropout() subcommand on the user created program Traj, I'm tying to use the command below, however, something strange happens, when I specify the whole age range (age_55-age_90) my stata crashes but when I truncate the age to age_55-age_76 the program works without issue, also if I truncate the command using the age rage age_80-age-90 the command also generates the proper model, it is only when the full age range age_55-age_90 is used that my Stata crashes.

traj, model(cnorm) var(cogtot27_55-cogtot27_90) order(1,3,3,1) indep(age_55-age_90) min(0) max(27) obsmar(death_indicator) weight(new_weight) dropout(00000)

[George Ford]

no one can replicate the issue without some data

[Luis Mijares Castaneda]

no one can replicate the issue without some data

I'm sorry for this but I tried using the dataex command but there are too many variables specified so I extracted a small number in this excell sheet

[Luis Mijares Castaneda]

no one can replicate the issue without some data

I generated a random sample from my data set to attach, this data set has 2,338 obs and I wanted to use the traj user generated command with the model shown below.

traj, model(cnorm) var(cogtot27_55-cogtot27_90) order(1,3,3,1) indep(age_55-age_90) min(0) max(27) obsmar(death_indicator) weight(new_weight) dropout(0,0,0,0)

[George Ford]

Yup. The crash can be replicated.


This is not Stata's code, but I might send this anyway to [email protected].