Re: st: left-side censoring of the independent variable

[Steve Samuels <sjsamuels@gmail.com>]

I should have added:  

If you know that risk started before your "pre-defined period" but you  observe events only after that period, and if, in addition, you know the date that risk started for all people you observe, then you can use the parametric regression models in-streg-.   Just specify the origin() and enter() options in -stset-. 

Steve




You can't solve it in "STATA" (see the last entry of the list FAQ for the reason why), but you can probably solve it in Stata . The phenomenon you describe is not left censoring, at least in the survival terminology that Stata uses, but left-truncation, and the pre-defined period is the truncation point.

See the Statalist thread beginning at: http://www.stata.com/statalist/archive/2011-04/msg01276.html
especially the Wooldridge reference given in http://www.stata.com/statalist/archive/2011-04/msg01324.html


For a solution in Stata, download -stregshared- from http://vandenberg.vwl.uni-mannheim.de/2999.0.html .   I haven't tried it so can't give any advice about it.



Steve
sjsamuels@gmail.com


On Jan 24, 2012, at 7:41 AM, massimiliano stacchini wrote:

Dear users,
I'm dealing with left-side censoring of the primary regressor in my empirical model (in more detail, the
duration of a phenomenon is my  independent variable and it may be observed onlysince a pre-defined period onwards).
How do you treat this problem with STATA ? 

Thanks 

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/