Suppose k events from a partition of the sample space Ω, i.e., they are disjoint and ∪ i=1 to k over Ai = Ω. Assume that P(B) > 0. Prove that if P(A1|B) < P(A1) then P(Ai | B) > P(Ai) for some i = 2, ... , k.

I used the law of total probability to prove P(A1|B) < P(A1) but I have no idea how to prove the that P(Ai | B) > P(Ai) for some i = 2, ... , k.

Thank you.

  • 2
    $\begingroup$ The law of total probability cannot be used to prove that $P(A_1\mid B) < P(A_1)$ (as you are claiming you did). If you do believe that you have a valid proof, then consider that the same proof can be adapted to prove that $P(A_i\mid B) < P(A_i)$ for all choices of $i$, not just $i=1$, and so you will have disproved what you set out to do. $\endgroup$ – Dilip Sarwate Apr 19 '15 at 20:31
  • $\begingroup$ That is why I am unable to understand how to prove that if for i=1 P(Ai | B) < P (Ai) then P(Ai | B) > P (Ai | B) for i=2, .. , k. Have you got any suggestions? $\endgroup$ – Octavian Apr 19 '15 at 20:41

Suppose it is not true.

Then $P(A_1\mid B)<P(A_1)$ and $P(A_i\mid B)\leq P(A_i)$ for $i=2,\dots,k$ leading to: $$1=P(\Omega\mid B)=\sum_{i=1}^kP(A_i\mid B)<\sum_{i=1}^kP(A_i)=P(\Omega)=1$$ A contradiction is found.

  • $\begingroup$ You answer is much appreciated! It makes sense now. $\endgroup$ – Octavian Apr 19 '15 at 21:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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