## With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

No credit card required

Problem 33

Problem. Consider the following games G1, G2, and G3, in each of which \$1 is won if a head is obtained, otherwise \$1 is lost (assuming the player starts with \$0).

G1: A biased coin that has probability of heads .495 is tossed.

G2: If the net gain is a multiple of three, coin A is tossed. The latter has probability of heads .095. If the net gain is not a multiple of three, coin B is tossed. The latter has probability of heads .745.

G3: G1 and G2 are played in any random order.

Prove that although G1 and G2 each result in a net expected loss, G3 results in a net expected gain.

Solution. For game G1 the net expected gain is \$1(.495) − \$1(.505) = −\$.01.

For game G2, let denote the net gain of the player after n tosses. Then, given all the prior values of the net gain, the value of depends only on , that is, is a Markov chain.1 Moreover, 2 is also a Markov chain with states {0, 1, 2} and transition probability matrix

After an infinitely large ...

## With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

No credit card required