Quote Originally Posted by CCFCC3PO View Post
We don't yet know how many people have contracted covid-19 without symptoms. The base rate needs to be an accurate figure.

Example.

x% of women aged 44-49 in South East Asia suffer from breast cancer. During testing of women aged 44-49 in South East Asia, the twice that percentage of women tested positive for breast cancer using a brand new method. Result - women are retested with the old method. First tests are found to have a significant number of false positives. Accuracy of test is calculated not just on false positives, but also based on that base rate.

The difference is, we know the base rate of women with cancer aged 44-49 in South East Asia. We don't know how many people have contracted covid. The other difference is that women are tested for breast cancer regardless of an assumption that they have it or not. In the main, testing of covid is done on people displaying symptoms.
I'm afraid you don't get it, you haven't answered my questions.
Are you saying that these medical modellers and statisticians across the World don't know how to weight samples ?