Sie sind auf Seite 1von 2

Indian Institute of Information Technology Allahabad

Probability & Statistics


C2 Review Test Marking Scheme

1. Let X be a constant random variable and Y be any random variable.


Are X and Y independent? Give proper justification. [4]
Solution: Suppose X = c, for some c ∈ R. The cdf of X is

1 if x ≥ c
FX (x) = [1]
0 x < c

Now the joint cdf of X and Y is

F(X,Y ) (x, y) = P (X ≤ x, Y ≤ y)

P (Y ≤ y) = FY (y) if x ≥ c, y ∈ R
= [2]
0 if x < c, y ∈ R

Clearly, F(X,Y ) (x, y) = FX (x)FY (y), for every (x, y) ∈ R2 . Thus X and
Y are independent. [1]
2. Consider a sequence of discrete random variables Yn with following prob-
ability mass function [5]

 1 − n1 y = 0,

1
P (Yn = y) = y = n2 ,
 n
0 otherwise.

Investigate the convergence of (Yn ) in probability.


Solution: Let  > 0 and n ∈ N. For  ≥ n2 , P (|Yn | ≥ ) = 1 − P (Yn <
) = 1 − P (Yn = 0) − P (Yn = n2 ) = 0. [2]
Now, for  < n2 , P (|Yn | ≥ ) = 1 − P (Yn < ) = 1 − P (Yn = 0) = n1 . [2]
Hence for every  > 0, lim P (|Yn | ≥ ) = 0. So (Yn ) converges to 0 in
n→∞
probability. [1]
3. Let X1 , X2 , X3 be independent identically distributed exponential ran-
dom variables with parameter 1. Find the joint probability density
function of
X 1 + X2 X1
Y1 = X1 + X2 + X3 , Y2 = , and Y3 = .
X1 + X2 + X 3 X1 + X2
Also find the marginal probability density function of Y1 , Y2 and Y3 . [11]
Solution: The joint pdf of X1 , X2 , X3 is given by

e−(x1 +x2 +x3 ) if 0 < x1 , x2 , x3 < ∞
f(X1 , X2 ,X3 ) (x1 , x2 , x3 ) = [1]
0 otherwise

x1 +x2 x1
Let y1 = x1 + x2 + x3 , y2 = x1 +x2 +x3 and y2 = x1 +x2 . Then we have
x1 = y1 y2 y3 , x2 = y1 y2 (1 − y3 ) and x3 = y1 (1 − y2 ) [ 21 + 12 + 21 ]
The jacobian of transformation is

∂x1 ∂x1 ∂x1 y y y1 y3

y1 y2
∂y1 ∂y2 ∂y3 2 3
J = ∂x
2 ∂x2 ∂x2
∂y1 ∂y1 ∂y1
= y2 (1 − y3 ) y1 (1 − y3 ) −y1 y2 = −y12 y2 [2]
∂x3 ∂x3 ∂x3
∂y ∂y ∂y y1 (1 − y2 ) y1 0
1 1 1

Thus the joint pdf of Y1 , Y2 , Y3 is given by

f(Y1 , Y2 ,Y3 ) (y1 , y2 , y3 ) = f(X1 , X2 ,X3 ) (y1 y2 y3 , y1 y2 (1 − y3 ), y1 (1 − y2 ))|J| [1]



y12 y2 e−y1 if 0 < y1 < ∞, 0 < y2 , y3 < 1
= [1+ 21 ]
0 otherwise

The marginal pdf of Y1 is given by


Z +∞ Z +∞
fY1 (y1 ) = f(Y1 , Y2 ,Y3 ) (y1 , y2 , y3 ) dy2 dy3 [1]
−∞ −∞
 12 y12 e−y1 if 0 < y1 < ∞
= [1]
0 otherwise

Similarly, the marginal pdf of Y2 is



2y2 if 0 < y2 < 1
fY2 (y2 ) = [1]
0 otherwise

The marginal pdf of Y3 is



1 if 0 < y3 < 1
fY3 (y3 ) = [1]
0 otherwise

Das könnte Ihnen auch gefallen