Show that $P(A|B)>P(A)$ if and only if $P(A^c |B)<P(A^c)$. We assume that $0<P(A)<1$ and $0<P(B)<1$ ( Finan #5.1.5)
Taking the complement of probabilities reverses the sign of the inequality (as shown below): $$P(X)>P(Y)\Rightarrow-P(X)<P(Y)\Rightarrow1-P(X)<1-P(Y)\Rightarrow P(X^c)<P(Y^c) $$
Hence, take the complement of both sides of $P(A|B)>P(A)$ giving us the desired result $$P(A^c |B)<P(A^c)$$
Is this correct?
Notes:
- Finan is a prep guide for Exam Probability of SOA-USA
- The answer key at the back of the book has a different, lengthier solution.
$P(A) = 1 - P(A^C)$
$P(A|B) = 1 - P(A^C|B)$
If $P(A|B) > P(A)$ then ...
$1 - P(A^C|B) > 1 - P(A^C)$
$-P(A^C|B) > -P(A^C)$
$P(A^C|B) < P(A^C)$