stsintreg

[Josh Schwartz]

I would like to use stsintreg (https://www.stata.com/manuals/ststintreg.pdf), but I don't understand how it works without stset. How does it know what your failure variable is or what the id variable is for multiple-record data?

[Clyde Schechter]

I have not used it, but as I understand the help file, the -interval()- option provides that information.

interval(t_l t_u) specifies two time variables that contain the endpoints of the censoring interval. t_l represents the lower endpoint, and t_u represents the
upper endpoint. interval() is required.

The interval time variables t_l and t_u should have the following form:

Type of data t_l t_u
--------------------------------------------------
uncensored data a = [a,a] a a
interval-censored data (a,b] a b
left-censored data (0,b] . b
left-censored data (0,b] 0 b
right-censored data [a,+inf) a .
missing . .
missing 0 .
--------------------------------------------------

And, as far as I know, it does not support multiple failure models, so there is no id variable.

By the way, it's -stintreg-, not -stsintreg-.

[Josh Schwartz]

Thanks for the response! So there's no way to use stintreg to analyze multiple-record data with time-varying covariates? It only has to be a single failure model