st: istdtize

[Javier Sanchez <jsanchez@upei.ca>]

I was trying to use the istdize command and realized that some of the 
outputs (eg. Number of observed events) are incorrect and change each 
time I run the command (this problem was posted before but could not 
find an answer). Also the number  of cases might be different from the 
same strata when looking at the stratified outputs as compared to the 
summary outputs. After exploring the ado file, I found that there might 
be a couple of lines missing in the file. The basic problem is when 
there is more than one group, the total number of cases is no added and 
the command is using a random line with the cases from that particular 
observation. Since I am not an expert on Stata, I added this basic lines 
to the ado file and the outputs look correct but it would be nice if 
somebody else could look into this as well since there are many papers 
citing the use of this command that might have some errors.

istdize.ado
line 1 ............
.
.
line 163 qui by `Gc': replace `topop'=`topop'[_N]
**************NEW*****************
       qui by `Gc': replace `cases'=sum(`cases')
       qui by `Gc': replace `cases'=`cases'[_N]
************** END NEW************



--
Javier Sanchez, DVM, PhD
Research Chair Risk Analysis
Atlantic Veterinary College
550 Univ. Ave
Charlottetown
C1A 4P3, PE
Canada

tel +1 902.566.0803
fax +1 902.566.0823

*
*   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/