Page 2 :
Scanned with CamScanner, (9-4), 62, : V(X/ Y=Y) = E(x²1 Y=Y) - [E(XIY=Y)], Simillarly swe can show that,

Page 3 :
Scanned with CamScanner, (18, Regression of X on Y and Reqression of Y on X :-, The conditional mean of x given Y:Y is,, E (x) Y•Y) e, suppose y= y-dz, .. E(XlY=Y) z M, t 8 y, which represents the line of regression of X on Y', simillarly, Conditional mean of Y given X-x i's,, ECY/X=x) = M2+8 2 CX-M), suppose x'c X-lz, E (Y/X=x) - Mz+.P 62 x', which represents the line of regregre. regress ion of Y on x., Moment Generating Function (M.G-F.) :-, Let CX,Y) ng BND(M,, Mz, Gi, 6, ) then, f (x,Y)= I, 2(1-p2), e, 2 TT 6, 62 V-82, Bi96270