08-09-2017 02:55 PM
I am trying to find how to get overall stratified Newcombe 95% confidence interval from FREQ, but I can only get CI for each stratum, not the overall. How do I get overall Newcombe CI?
From here, http://support.sas.com/documentation/cdl/en/procstat/67528/HTML/default/viewer.htm#procstat_freq_det..., (the last few paragraph) looks like I can get Wald CI for each stratum first, and then have to manually combine each stratum Wald CI by using Mantel-Haenszel weight to get common Wald CI, and then use common Wald CI to calculate common Newcombe CI. Please help!