Course Title : Probability II
Instructor : Dr. ROUABAH
2024–2025
HomeWork No. 1
Problem. (A derivative of the bivariate normal distribution)
Let Z1 and Z2 be independent N (0, 1) r.vs., and define new r.vs. X and Y by
X = aX Z1 + bX Z2 + cX and Y = aY Z1 + bY Z2 + cY ,
where aX , bX , cX , aY , bY and cY are constants.
1. Show that
E(X) = cX , Var(X) = a2X + b2X ,
E(Y ) = cY , Var(Y ) = a2Y + b2Y ,
Cov(X, Y ) = aX aY + bX bY .
2. If we define the constants aX , bX , cX , aY , bY and cY by
s s
1+ρ 1−ρ
aX = σX , bX = σX , cX = µX ,
2 2
s s
1+ρ 1−ρ
aY = σY , bY = - σY , cY = µY ,
2 2
2
where µX , µX , σX , σY2 and ρ are constants, −1 ≤ ρ ≤ 1, then show that
2
E(X) = µX , Var(X) = σX ,
E(Y ) = µY , Var(Y ) = σY2 ,
ρXY = ρ.
2
3. Show that (X, Y ) has the bivariate normal pdf with parameters µX , µY , σX , σY2 and ρ.
2
4. If we stars with bivariate normal parameters µX , µY , σX , σY2 and ρ, we can define
constants aX , bX , cX , aY , bY and cY as the solutions to the equations
2
µ X = cX , σX = a2X + b2X ,
µ Y = cY , σY2 = a2Y + b2Y ,
ρσX σY = aX aY + bX bY .
Show that the solution given in part (2) is not unique by exhibiting another solution to
these equations. How many solutions are there ?
