Dear stata users,

I am conduction research after risk factors for mortality in thoracic trauma. I am using logistical regression with a backward stepwise method. I have a final model with 469 observations and stata automatically deletes some observations because of estimability and missing data. I now want to label the observations that are included in the model that I am interested in order to be able to investigate the characteristics of the observations that are excluded from my model. Is there a way to generate a new variable that will label the observations that are included in the model that I am interested in?

the current code I am using for my logistical regression is

* model 29
xi : stepwise, pr(.157) : logistic died (i.gender) triss_09 (i.cci)(i.cps)
estat gof, group(10)

with variables
died(whether the patient died or not)
gender 0 1
triss_09 -> value that predicts survival risk depending on the severity of the injury and physiological parameters.
CCI: comprehensive complication index: a weighted value for the total of complications that occured.
CPS: comorbidity polypharmacy score: the total number of medications and comorbidities a patient has.

Best regards,

Anne Fokkema, the Netherlands.