Reviewer of our article has required to indicate "exact 95CI" for frequencies of different events in cohort. We use Clopper-Pearson nethod for binomial proportions but which one is suitable for multinomial ones? Reviewer wrote "calculate as in StatXact soft", and "use similar methods" (which ones, aurhors were not indicated) . We downloaded trial version of StatXact and it works but it is too expensive. Even more, we have academic StatsDirect and SPSS already and I sure it is enough, but cannot find there such test for proportions. May be you know which nethod is used in StatXact and where it is hidden in SPSS/PAST? Or help with free soft or online calc. Thank you.
Dear Aleksandr Matveev ,
Hi,
You can try NTSYS-pc, a user-friendly and easy to use with excellent visual effect and editable graph if you want to give it a particular look.
Best,
Saeed
You can see a sample of regression with confident ellipse drawn by NTSYS-pc in my article, see link below
Article Gathering, Identification, Medicinal Utilization and Domesti...
Best,
Saeed
This may help: https://blogs.sas.com/content/iml/2017/02/15/confidence-intervals-multinomial-proportions.html
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS