Use Theorem 4.14 and its corollary to show that if X11, X12,...,X1n1,X21,X22,...,X2n2 are independent random variables, with the first n1 constituting a random sample from an infinite population with the mean mu1 and the variance sigma 1 squared and the other n2 constituting a random sample from an infinite population with the mean mu2 and the variance sigma2 squared, then a)E(X1-X2) (both X's have bars on top) = mu1-mu2; b) var(X1-X2) (both X's have bars on top) = sigma1 squared divided by n1 plus sigma2 squared divided by n2 Theorem 4.14: If X1, X2...Xn are random variables and Y=the sum of aiXi i=1 where a1, a2, and an are constants, then E(Y)= the sum of ai E(Xi) i=1 and var(Y) = the sum of ai squared times var(Xi)+2 sigma sigma aiaj times cov (XiXj) where the double summation extends over all values of i and j, from 1 to n, which i< Thanks! possible. if before him used I as again Pillai Rajeevan use to like would> π1 π1 π1 π=1 π=1 π=1 π2 π2 π2 π=1 π=1 π=1 1 ∑ππ=1 π1π 1 1 1 Μ πΈ(π1 ) = πΈ ( ) = πΈ (∑ π1π ) = ∑ πΈ(π1π ) = ∑ πΈ(π1π ) π1 π1 π1 π1 π1 = 1 π1 π1 = π1 ∑ π1 = π1 π1 π=1 2 ∑ππ=1 π2π 1 1 1 πΈ(πΜ 2 ) = πΈ ( ) = πΈ (∑ π2π ) = ∑ πΈ(π2π ) = ∑ πΈ(π2π ) π2 π2 π2 π2 π2 = 1 π2 π2 = π2 ∑ π2 = π2 π2 π=1 πΈ(πΜ 1 − πΜ 2 ) = πΈ(πΜ 1 ) − πΈ(πΜ 2 ) = π1 − π2 π1 π1 π1 π1 π=1 π=1 π=1 π=1 1 1 2 1 1 π(πΜ 1 ) = π (∑ π1π ) = ∑ ( ) π(π1π ) + 2 ∑ ∑ ( ) ( ) πΆππ£(ππ , ππ ) π1 π1 π1 π1 π<π π1 π1 π1 π=1 π=1 π=1 1 2 2 1 1 1 2 π1 2 2 = ∑ ( ) π1 + 2 ∑ ∑ ( ) ( )(0) = ( ) (π1 π1 ) + 0 = π1 π1 π1 π1 π1 π<π (πππππ π‘βπ π£ππππππππ π1π πππ π1π πππ πππππππππππ‘ πππ£(π1π , π1π ) = 0 ) π2 π2 π2 π2 π=1 π=1 π=1 π=1 1 1 2 1 1 π(πΜ 2 ) = π (∑ π2π ) = ∑ ( ) π(π2π ) + 2 ∑ ∑ ( ) ( ) πΆππ£(ππ , ππ ) π2 π2 π2 π2 π<π π2 π2 π2 π=1 π=2 π=1 1 2 2 1 1 1 2 π2 2 2 = ∑ ( ) π1 + 2 ∑ ∑ ( ) ( )(0) = ( ) (π2 π2 ) + 0 = π2 π2 π2 π2 π2 π<π (πππππ π‘βπ π£ππππππππ π2π πππ π2π πππ πππππππππππ‘ πππ£(π2π , π2π ) = 0 ) πΈ(πΜ 1 − πΜ 2 ) = (1)2 π(πΜ 1 ) + (−1)2 π(πΜ 2 ) + 2(1)(−1)πΆππ£(πΜ 1 , πΜ 2 ) = π(πΜ 1 ) + π(πΜ 2 ) + 2(1)(−1)(0) πππππ πΜ 1 πππππππ ππππ¦ ππ π1π ′π πππ πΜ 2 πππππππ ππππ¦ ππ π2π ′π (Μ ) π1 πππ πΜ 2 πππ πππππππππππ‘ πππ π‘βπππππππ π‘βπππ πππ£πππππππ ππ 0 π1 2 π1 2 = π(πΜ 1 ) + π(πΜ 2 ) = + π1 π2