Given a data-set like below, how to find the probability of survival given specific age, sex & disease condition, once someone gets infected?

Fatality_rate(Age 0-19) = 0.2
Fatality_rate(Age 20-19) = 0.4

Fatality_rate(Male) = 2.8
Fatality_rate(female) = 1.7

Fatality_rate(Hypertension) =   2.8
Fatality_rate(Cardiovascular_diseases) = 1.7

For example, how could I calculate the Fatality_rate of a Male with Hypertension of age 40 ?

Fatality_rate = Total-Deaths-of-Total-Infected-Confirmend*100/ Total-Infected-Confirmend
  • $\begingroup$ Given your definition, it can't be >1, so you mean % in here? $\endgroup$
    – Tim
    Commented Mar 2, 2020 at 6:34
  • $\begingroup$ You are right, have updated the formula accordingly. $\endgroup$
    – pauljeba
    Commented Mar 2, 2020 at 8:10

1 Answer 1


You can't. By the definition of Independence

$$ p(A,B,C) = p(A)\,p(B)\,p(C) $$

so you could calculate it if you could assume Independence, but you can't. Obviously things like hypertension are related to age. By assuming independence, you'd be saying that teenage female has same chance of hypertension as 60 year old male, what is incorrect. If you can't assume independence, you'd need to know joint distribution, while you don't.

  • $\begingroup$ exactly, they are not independent. So I am doomed without having access to the joint distribution, which I don't. $\endgroup$
    – pauljeba
    Commented Mar 2, 2020 at 8:09
  • $\begingroup$ @pauljeba then you can't get the probabilities from them. You'd need additional information for that. Otherwise, assuming independence could give you a really bad approximation. $\endgroup$
    – Tim
    Commented Mar 2, 2020 at 8:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.