FS#11 - TAB does not calculate OR for large N
gen i ill=ran(2)
gen i expose=ran(2)
tab ill expose /or // get OR=0, CI=0,0 <--*******
tab ill expose /rr // get correct RR (close to 1)
The value of N is greater than 6000, but with my random numbers the point at which OR is not calculated varies (e.g. wrong at n=6400, OK at n=6420) so it must also have to do with the distribution of values in the table.
This will be easy to add to a test script