Here is a new proof problem: We have seen the law of total probability and the law of total expectation and we like the movie Total Recall, so lets prove the law of total variance. It goes as follows: Var(X) = E(var(X|Y)) + var(E(X|Y)). (a) First show that E(E(X|Y)) = E(X) (b) Explain why the above is true. (c) Use the above fact and show that Var(X) = E(var(X|Y)) + var(E(X|Y)) Solution: var(X) = E(X^2) - E(X)^2 = E(E(X^2|Y)) - E(E(X|Y))^2 = E(var(X|Y)) + E(E(X|Y)^2) - E(E(X|Y))^2 = E(var(X|Y)) + var(E(X|Y)). The following 3 are from CPs and PSets: cp4f Problem 4. Extra Problem. (a) Exhibit three nonisomorphic, connected graphs with five vertices and four edges. (b) Argue that every connected graph with five vertices and four edges is isomomorphic to one of the three in part (a). cp14w Problem 5. The covariance, Cov [X, Y ], of two random variables, X and Y , is defined to be E [XY ]- E [X] E [Y ]. Note that if two random variables are independent, then their covariance is zero. (a) Give an example to show that having Cov [X, Y ] = 0 does not necessarily mean that X and Y are independent. (b) Let X1, . . . ,Xn be random variables. Prove that Var [X1 + · · · + Xn] = Sum i=1..n Var [Xi] + 2 Sum i