Originally Posted by
wphamilton
Yeh, the 14% of everyone having heart failure and the 97% NY fatalities are both too broad to be useful. As I mentioned before, when you have more information you can use things like Baye's Theorem to get probabilities that are more specific.
Overbroad or not, do you still claim that the 14% percent is predictive for individual elements within the population, or are you going to sweep that under the carpet?