Am trying to using a forward stepwise model building approach to
progressively include covariates in the model as below
xi: stcox i.currentagegrp i.poor i.babies_bor i.pct_agegrp, robust
cluster(rsv)
est store a1
xi: stcox i.currentagegrp i.poor i.babies_bor, robust cluster(rsv)
est store b1
lrtest a1 b1
is it ok to use the lrtest to see which model fits best to check
whether its necessary to include the variable for pct_age as a possible
risk factor. The issue being that Cox model uses a partial likelihood
thanks
Emelda
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
