Answer to Question #215091 in Statistics and Probability for Arshad Ali Wazir

Question #215091

Q.4 A random vector (X, Y, Z) has joint density given by 

. f (x, y,z) = k exp [− 1/2 (2x2 − 2xy + y2 + z2 + 2x − 6y) . 

1. Compute k.

2. Compute the expectations P(X), P(Y ) and P(Z).

3. Compute the density of the random vector (X, Z).

4. Compute the correlation coeffificient between X and Z and between X and Y . 

5. Let W = X + Z; compute the probability density of W. 



1
Expert's answer
2021-07-16T09:14:53-0400

"f (x, y, z) = k \\exp [\u2212 (2x^2 \u2212 2xy + y^2 + z^2 + 2x \u2212 6y)\/2]"


"f (x, y, z) = k \\exp [\u2212 ((x-2)^2 \u2212 (y-x-3)^2 + z^2 -13)\/2]"


"f (x, y, z) = k \\exp [\u2212 (2(x-(y-1)\/2)^2 + (y-5)^2\/2 + z^2 -13)\/2]"


(1) "1=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}f(x,y,z)dxdydz"


"=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}k \\exp [\u2212 ((x-2)^2 \u2212 (y-x-3)^2 + z^2 -13)\/2]dxdydz"


"=ke^{-13\/2}\\int\\limits_{-\\infty}^{+\\infty}\\exp(- z^2\/2)dz\\int\\limits_{-\\infty}^{+\\infty}\\exp (\u2212 (x-2)^2\/2)\\left(\\int\\limits_{-\\infty}^{+\\infty}\\exp(\u2212(y-x-3)^2 \/2)dy\\right)dx"

"=k(2\\pi)^{3\/2}e^{-13\/2}"

Therefore, "k=(2\\pi)^{-3\/2}e^{13\/2}".


(2) "E(X)=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}xf(x,y,z)dxdydz="

"=ke^{-13\/2}\\int\\limits_{-\\infty}^{+\\infty}\\exp(- z^2\/2)dz\\int\\limits_{-\\infty}^{+\\infty}x\\exp (\u2212 (x-2)^2\/2)\\left(\\int\\limits_{-\\infty}^{+\\infty}\\exp(\u2212(y-x-3)^2 \/2)dy\\right)dx="

"=\\frac{1}{\\sqrt{2\\pi}}\\int\\limits_{-\\infty}^{+\\infty}x\\exp (\u2212 (x-2)^2\/2)dx=2"


"E(Y)=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}yf(x,y,z)dxdydz="

"=ke^{-13\/2}\\int\\limits_{-\\infty}^{+\\infty}\\exp(- z^2\/2)dz\\int\\limits_{-\\infty}^{+\\infty}\\exp (\u2212 (x-2)^2\/2)\\left(\\int\\limits_{-\\infty}^{+\\infty}y\\exp(\u2212(y-x-3)^2 \/2)dy\\right)dx="

"=\\frac{1}{\\sqrt{2\\pi}}\\int\\limits_{-\\infty}^{+\\infty}(x+3)\\exp (\u2212 (x-2)^2\/2)dx=5"


"E(Z)=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}zf(x,y,z)dxdydz="

"=ke^{-13\/2}\\int\\limits_{-\\infty}^{+\\infty}z\\exp(- z^2\/2)dz\\int\\limits_{-\\infty}^{+\\infty}\\exp (\u2212(x-2)^2\/2)\\left(\\int\\limits_{-\\infty}^{+\\infty}\\exp(\u2212(y-x-3)^2 \/2)dy\\right)dx="

"=0"


(3)

"f_{X,Z} (x, z)=\\int\\limits_{-\\infty}^{+\\infty}f(x,y,z)dy=\\int\\limits_{-\\infty}^{+\\infty} k \\exp [\u2212 ((x-2)^2 \u2212 (y-x-3)^2 + z^2 -13)\/2]dy="

"=\\frac{x+3}{2\\pi}\\exp [\u2212 ((x-2)^2 + z^2 )\/2]"


"V(X)=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}(x-2)^2f(x,y,z)dxdydz="

"=ke^{-13\/2}\\int\\limits_{-\\infty}^{+\\infty}e^{- z^2\/2}dz\\int\\limits_{-\\infty}^{+\\infty}(x-2)^2 e^{\u2212(x-2)^2\/2} \\left(\\int\\limits_{-\\infty}^{+\\infty}e^{\u2212(y-x-3)^2 \/2}dy\\right)dx=1"


"V(Y)=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}(y-5)^2f(x,y,z)dxdydz="

"=ke^{-13\/2}\\int\\limits_{-\\infty}^{+\\infty}e^{- z^2\/2}dz\\int\\limits_{-\\infty}^{+\\infty} e^{\u2212(x-2)^2\/2} \\left(\\int\\limits_{-\\infty}^{+\\infty}(y-5)^2e^{\u2212(y-x-3)^2 \/2}dy\\right)dx="

"=\\frac{1}{2\\pi}\\int\\limits_{-\\infty}^{+\\infty} e^{\u2212(x-2)^2\/2} \\left(\\int\\limits_{-\\infty}^{+\\infty}(y+x-2)^2e^{\u2212y^2 \/2}dy\\right)dx="

"=\\frac{1}{\\sqrt{2\\pi}}\\int\\limits_{-\\infty}^{+\\infty} (1+(x-2)^2)e^{\u2212(x-2)^2\/2}dx="

"=\\frac{1}{\\sqrt{2\\pi}}\\int\\limits_{-\\infty}^{+\\infty} (1+x^2)e^{\u2212x^2\/2}dx=2"


(4) "Cov(X, Z)=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}z(x-2)f(x,y,z)dxdydz="

"=ke^{-13\/2}\\int\\limits_{-\\infty}^{+\\infty}ze^{- z^2\/2}dz\\int\\limits_{-\\infty}^{+\\infty}(x-2) e^{\u2212(x-2)^2\/2} \\left(\\int\\limits_{-\\infty}^{+\\infty}e^{\u2212(y-x-3)^2 \/2}dy\\right)dx=0"


Therefore, "Corr(X,Z)=Cov(X,Z)\/(\\sigma_X\\sigma_Z)=0"


"Cov(X, Y)=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}(x-2)(y-5)f(x,y,z)dxdydz="

"=ke^{-13\/2}\\int\\limits_{-\\infty}^{+\\infty}e^{- z^2\/2}dz\\int\\limits_{-\\infty}^{+\\infty}(x-2) e^{\u2212(x-2)^2\/2} \\left(\\int\\limits_{-\\infty}^{+\\infty}(y-5)e^{\u2212(y-x-3)^2 \/2}dy\\right)dx="

"=\\frac{1}{2\\pi}\\int\\limits_{-\\infty}^{+\\infty}(x-2) e^{\u2212(x-2)^2\/2} \\left(\\int\\limits_{-\\infty}^{+\\infty}(y+x-2)e^{\u2212y^2 \/2}dy\\right)dx="

"=\\frac{1}{\\sqrt{2\\pi}}\\int\\limits_{-\\infty}^{+\\infty}(x-2)^2 e^{\u2212(x-2)^2\/2}dx=1"


"\\sigma_X=\\sqrt{V(X)}=1"

"\\sigma_Y=\\sqrt{V(Y)}=\\sqrt{2}"


"Cov(X, Y)=Corr(X,Y)\/(\\sigma_X\\sigma_Y)=1\/(1\\cdot\\sqrt{2})=\\sqrt{2}\/2"


5."W=X+Z"

"P(W<w)=\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}\\int\\limits_{-\\infty}^{+\\infty}I_{x+z<w}(x,y,z)f(x,y,z)dxdydz="

"=ke^{-13\/2}\\int\\limits_{-\\infty}^{+\\infty}e^{- z^2\/2}\\int\\limits_{-\\infty}^{w-z}e^{\u2212(x-2)^2\/2} \\int\\limits_{-\\infty}^{+\\infty}e^{\u2212(y-x-3)^2 \/2}dydxdz="

"=\\frac{1}{2\\pi}\\int\\limits_{-\\infty}^{+\\infty}e^{- z^2\/2}\\int\\limits_{-\\infty}^{w-z}e^{\u2212(x-2)^2\/2} dxdz"


"f_W(w)=\\frac{d}{dw}P(W<w)=\\frac{1}{2\\pi}\\frac{d}{dw}\\int\\limits_{-\\infty}^{+\\infty}e^{- z^2\/2}\\int\\limits_{-\\infty}^{w-z}e^{\u2212(x-2)^2\/2} dxdz="

"=\\frac{1}{2\\pi}\\int\\limits_{-\\infty}^{+\\infty}e^{- z^2\/2}e^{\u2212(w-z-2)^2\/2}dz="

"=\\frac{1}{2\\pi}\\int\\limits_{-\\infty}^{+\\infty}e^{-(2z^2-2zw+w^2+4z-4w+4)\/2}dz="

"=\\frac{1}{2\\pi}\\int\\limits_{-\\infty}^{+\\infty}e^{-(z-w\/2+1)^2}e^{-(w-2)^2\/4}dz="

"=\\frac{1}{2\\pi}e^{-(w-2)^2\/4}\\int\\limits_{-\\infty}^{+\\infty}e^{-z^2}dz=\\frac{1}{2\\sqrt{\\pi}}e^{-(w-2)^2\/4}"

Need a fast expert's response?

Submit order

and get a quick answer at the best price

for any assignment or question with DETAILED EXPLANATIONS!

Comments

No comments. Be the first!

Leave a comment

LATEST TUTORIALS
New on Blog