Assignment 1

Q1)
We have given pdf of x
1
𝑓 ( 𝑥 )= , 0 ≤ 𝑥 ≤ 𝑘
𝑘

The three different methods of obtaining a simulated

value of X are
𝑥 1 ​=𝑟 ⋅ 𝑘

𝑥 2 ​=𝑘⋅ ( 1 −𝑟 )

𝑥 3=𝑘𝑟

Where r is a random number generated from a uniform

distribution over [0,1]
i) x1 =r⋅k:
Since r is uniformly distributed over [0,1], multiplying r by
k results in values of x1 that are uniformly distributed
over the interval [0,k]
This is a valid simulation, as x1 lies within the support of
X, which is [0,k].
ii)x2 =k⋅(1−r):
For r∈[0,1] , 1−r will also be in the range [0,1].
Multiplying k by 1−r gives values of x that are uniformly
distributed over [0,k], because the transformation is
essentially a scaling of the uniform random variable 1−r.
This is also a valid simulation, as x2 lies within the
support of X, which is [0,k].
iii) x3 =k/r :
Since r∈[0,1] , dividing by k results in x3∈[0,1/k].
However, X is uniformly distributed over [0,k], so x3 lies
outside the support of X (which is [0,k]).
This is not a valid simulation, because the values of xx3
do not match the range of possible values for X
Q2)
i)For binomial distribution
𝑋 𝐵𝑖𝑛 ( 𝑛 , 𝑝 )

MEAN = np = 10
10
n= 𝑝

Variance = np(1-p) = 5
10 −10 𝑝=5

P=0.5
=> n=20
𝑋 𝐵𝑖𝑛 ( 20 , 0.5 )

𝑃 ( 𝑋 ≤ 𝑥 )=0.0207

From the tables of binomial distribution the value of x is

X=5

ii) For normal distribution

𝑋 ∼ 𝑁 (𝜇 , 𝜎2)

Mean= 𝜇 = 0
Variance=𝜎 =1 2
 , we can say that x follows standard normal
𝑋 ∼ 𝑁 ( 0 , 1)

distribution as mean and variance are 0 and 1

respectively.
𝑃 ( 𝑋 ≤ 𝑥 )=0.1190

From the table of normal distributions we can say that no

value of x can be find as in tables values for 𝜙(x) starts
from 0.5 itself which is greater than 0.1190.

iii) for chi-square distribution

2
𝑋 𝜒𝑣

Variance= 2v = 72 => v=36

So, mean = v = 36
X=51
𝑃 ( 𝑋 ≤51 ) =𝐹 ( 51 )
2
𝑋 ∼ 𝜒 36

𝑃 ( 𝜒 236 ≤ 51 ) ≈ 0.9999
Q3) & Q5) Given
4
i) f(x)= 5 𝑥+1
, 𝑥=0.1 .2 … .

∞
𝑀 𝑥 ( 𝑡 )=∑ 𝑒 𝑡𝑥 𝑓 ( 𝑥 )
𝑥=0

∞
4
𝑀 𝑥 ( 𝑡 )=∑ 𝑒 𝑡𝑥 𝑥 +1
𝑥=0 5

( 45 )+(𝑒 54 )+( 𝑒 54 )+… … … … . .

0
¿ 𝑒
𝑡
2
2𝑡
3

[ ]
𝑡 2𝑡
1 𝑒 𝑒
¿4 + + +…….
5 5 2 53

𝑎
Sum of infinite gp Sn = 1− 𝑟

[ ]
1
5
¿4
( 5 −𝑒𝑡 )
5

¿4
[( )]
5− 𝑒𝑡
1

Hence Mx (t) =¿ 4 [ 5− 𝑒 ] , 𝑓𝑜𝑟 𝑒 <5 𝑡 −1 𝑡

ii) 𝐸 ( 𝑥 ) =𝑀 ❑ ( 0 ) ′
𝑥

𝑡
′ 4𝑒
𝑀 ❑𝑥 ( 𝑡 )=− 2
( 5 −𝑒 𝑡 )
0
( ) 4𝑒
′ 0
𝑀❑ 0 = 𝑥 2
, 𝑒 =1
( 𝑒 −5 )
0

′
𝑀 ❑ ❑𝑥 ( 𝑡 )=0.25
Q4)
❑

i) 𝑀 ❑𝑥 ( 𝑡 )=∫ 𝑒𝑡𝑥 𝑓 ( 𝑥 ) 𝑑𝑥
𝑥

[ ][ ]
0 ∞
1 𝑥 𝑡𝑥 1 − 𝑥 𝑡𝑥
= 𝑀 ❑𝑥 ( 𝑡 )= ∫ 𝑒 𝑒 𝑑𝑥 + ∫ 𝑒 𝑒 𝑑𝑥
−∞ 2 0 2

[∫ ] [∫ ]
0 ∞
1 𝑥 (1+𝑡 ) 1 𝑥 ( 𝑡 − 1)
= 𝑀 ❑𝑥 ( 𝑡 )=
−∞ 2
𝑒 𝑑𝑥 +
0 2
𝑒 𝑑𝑥

= {( ) ( )}
0 0
1 𝑒 𝑒
+ −
2 ( 1+𝑡 ) ( 𝑡 −1 )

𝑀 𝑥 ( 𝑡 )=
1
( 2
2 ( 1 −𝑡 2 ) )
Hence, 𝑀 𝑥 ( 𝑡 )=( 1 −𝑡 2 )
−1
for |t|¿1
 ii) 𝐸 ( 𝑋 )=𝑀 ❑ ( 0 ) ′
𝑥

❑
𝑑
𝑀 ❑′𝑥 ( 𝑡 )= ❑ ( 𝑀𝑥 𝑡 )
( )
𝑑𝑥
2
𝑉𝑎𝑟 ( 𝑋 )=𝑀 ❑ ❑𝑥 ( 0 ) − ( 𝑀 ❑𝑥 (𝑡 ) )
′ ′ ′

′ 2𝑡
𝑀 ❑ ❑𝑥 ( 𝑡 )= 2
( 1 −𝑡 2 )
′
𝑀 ❑ ❑𝑥 ( 0 )=𝐸 ( 𝑋 )=0

Hence the value of mean is 0

2
𝑑
Now ( )
𝑀❑ 𝑡 =
𝑑𝑥
2(
′′
𝑥 𝑀 𝑥 ( 𝑡 ))

′ ′
𝑀 ❑ ❑𝑥 ( 𝑡 )=
[ ( 1 −𝑡 2) 2 −2 ( 1 −𝑡2 ) ( −2 𝑡 ] ]
4
( 1− 𝑡 2 )
′ ′
𝑀 ❑ ❑𝑥 ( 0 )=2
𝑉𝑎𝑟 ( 𝑋 )=2

Hence the value of variance is 2

Q6)
i) A model where the standard deviation is greater
than the mean means that the claims are highly
predictable. Most claims might be small, but there
are chances of very large claims happening making
the data spread out widely.

ii) a)
We have cdf
( )
5
10
𝐹 ( 𝑥 )=1−
( 10+ 𝑥 )

Let, F(x) = r
As we know,
If 𝑈 𝑈𝑛𝑖𝑓𝑜𝑟𝑚 ( 0 , 1 ) then,
−1
𝐹 ( 𝑈 )= 𝑋

So,
( )
5
10
𝑟 =1−
( 10+ 𝑥 )
1
10 5
=( 1 −𝑟 )
10+ 𝑥

𝑟 =10
[ 1
1
−1
( 1 −𝑟 )5 ]
𝑥=10 [ ( 1− 𝑟 )−0.2 −1 ]

Hence the required transformation to generate a

simulated value of X is:
𝑥=10 [ ( 1− 𝑟 )−0.2 −1 ]
b)
It is given that
𝑟 𝑈 ( 0 , 1 ) , r is following uniform distribution so we can

also say that

1 −𝑟 𝑈 ( 0 , 1 ) , 1 – r will also follow normal distribution

If 𝑟 𝑈 ( 0 , 1 )
=> 𝑥=10 [ ( 1− 𝑟 ) −1 ]
−0.2

Then,
1 −𝑟 𝑈 ( 0 , 1 )

=> 𝑥=10 [ 𝑟 −0.2

−1 ]

c)
𝑟 =0.0016 (given)
𝑥=10 [ ( 𝑟 ) −1 ]
−0.2

Putting the value of r = 0.0016 in above equation

𝑥=10 [ ( 0.0016 )− 0.2 − 1 ]
𝑥=2 ,62,389

Now ,
𝑟 =0.5154
𝑥=10 ( ( 0.5154 )−0.2 −1 )
𝑥=14,175

Hence , the missing frequencies are x=2,62,389 and

x=14,175 respectively.
Q7)
i) We have given joint probability distribution
function
− ( 𝑥 +𝑦 )
𝑓 𝑥 , 𝑦 ( 𝑥 , 𝑦 )=𝑒 , 𝑥> 0 , 𝑦 >0

For marginal density function of x and y ;

❑
𝑓 𝑥 ( 𝑥 )=∫ ❑ 𝑓 𝑥 , 𝑦 ( 𝑥 , 𝑦 ) 𝑑𝑦
𝑦

❑
𝑓 𝑦 ( 𝑦 )=∫ ❑ 𝑓 𝑥 , 𝑦 ( 𝑥 , 𝑦 ) 𝑑𝑥
𝑥

∞
𝑓 𝑥 ( 𝑥 )=∫ ❑𝑒
− ( 𝑥+ 𝑦 )
𝑑𝑦
0
∞
𝑓 𝑦 ( 𝑦 )=∫ ❑ 𝑒
− ( 𝑥+ 𝑦 )
𝑑𝑥
0

− ( 𝑥+ 𝑦 ) ∞
𝑓 𝑥 ( 𝑥 )=− [ 𝑒 0 ]
− ( 𝑥+ 𝑦 ) ∞
𝑓 𝑦 ( 𝑦 )=− [ 𝑒 0 ]

On solving we get ,
−𝑥
𝑓 𝑥 ( 𝑥 )=− 𝑒 , x>0
𝑓 𝑦 ( 𝑦 )=− 𝑒
−𝑦
, y>0
To check whether variables are independent
𝑓 𝑥𝑦 ( 𝑥 , 𝑦 )= 𝑓 𝑥 ( 𝑥 ) 𝑓 𝑦 ( 𝑦 )

¿ 𝑒 −𝑥 𝑒− 𝑦 ⇒ 𝑒− (𝑥 +𝑦 )

Hence x and y are independent as LHS = RHS

ii) Joint cumulative distribution function
[∫ ]
𝑥 𝑦
𝐹𝑥𝑦 ( 𝑥 , 𝑦 )=∫ 𝑓𝑥𝑦 ( 𝑥 , 𝑦 ) 𝑑𝑦 𝑑𝑥
0 0
𝑦 𝑥
𝐹𝑥𝑦 ( 𝑥 , 𝑦 )=∫ ∫ 𝑒
− ( 𝑥+ 𝑦 )
𝑑𝑥𝑑𝑦
0 0

On solving we get
𝐹𝑥𝑦 ( 𝑥 , 𝑦 )=( 1− 𝑒− 𝑥 ) ( 1− 𝑒− 𝑦 )
 ❑
𝑃 ( 𝑋 =𝑥 , 𝑌 =2 )
Q8) 𝐸 [ 𝑋 /𝑌 =2 ] =∑ ❑ 𝑋
𝑥 𝑃 ( 𝑌 =2 )
𝑃 ( 𝑋 =0 , 𝑌=2 ) 𝑃 ( 𝑋 =1 ,𝑌 =2 ) 𝑃 ( 𝑋 =2, 𝑌 =2 )
𝐸 [ 𝑋 /𝑌 =2 ] =0 ⋅ +1 ⋅ +2 ⋅
𝑃 ( 𝑌 =2 ) 𝑃 ( 𝑌 =2 ) 𝑃 ( 𝑌 =2 )

𝑃 (𝑌 =2 )=0.40
0.15 0.20
= 0+ +
0.40 0.40
𝐸 [ 𝑋 /𝑌 =2 ] =0.875
Q9)
[𝑛 1 𝑥 1+𝑛 2 𝑥 2 ]
𝑋=
𝑛 1+𝑛 2
[ 25 ⋅120.2+18 ⋅142.7 ]
𝑋=
25+18

Hence the combined mean is

𝑋=129.6186
To calculate combined variance
❑

𝜎
2
=
∑ 𝑋 21 𝑖
−𝑥
2
❑ and
1 1
𝑛1
❑

2 ∑ 𝑋 2𝑖
2
2
𝜎 = 2 −𝑥 ❑ 2
𝑛2
2 2
𝜎 1 =58.1
2 2
𝜎 2=62.2

On calculating
❑
∑ 𝑋 21 𝑖=44591.25
❑
∑ 𝑋 22 𝑖=436178.34
❑ ❑
∑ 1 𝑖 ∑ 𝑋 22 𝑖 =480769.59
𝑋
2
+

Now using ,
❑
1
2❑
𝜎 =
𝑛1 +𝑛2
∑ 2
𝑋 𝑖 − 𝑥❑
2

2
𝜎 =3600.93
Hence combined standard deviation is
𝜎 =60.01
 Q10)
i) To show for continuous random variable X and Y
𝐸 ( 𝑌 )=𝐸 ( 𝐸 ( 𝑌 / 𝑋 ) )

We know that
❑
𝐸 ( 𝑋 )=∫ 𝑋 . 𝑓 ( 𝑥 ) 𝑑𝑥

So,
( ( )) ( )
❑
𝑌 𝑌
𝐸 𝐸 =∫ 𝐸 . 𝑓 ( 𝑥 ) 𝑑𝑥
𝑋 𝑋
❑
𝐸 ( )=∫ 𝑌 . 𝑓 ( ) 𝑑𝑦
𝑌 𝑦
𝑋 𝑥

[ ( 𝑥) ]
❑
𝐸 ( 𝐸 ( 𝑌 / 𝑋 ) ) =∫ ❑ ∫ 𝑦 𝑓 𝑦 𝑑𝑦 𝑓 ( 𝑥 ) 𝑑𝑥
𝑥 𝑦

𝑓 ( )=
𝑦 𝑓 (𝑥 , 𝑦 )
𝑥 𝑓 (𝑥)
𝑦
∫𝑦

𝐸 ( 𝐸 ( 𝑌 / 𝑋 ) ) =∫ 𝑓 ( 𝑥 , 𝑦 ) 𝑑𝑥 𝑑𝑦=𝐸 ( 𝑌 )
𝑥

Hence,
𝐸 ( 𝐸 ( 𝑌 / 𝑋 ) ) =𝐸 ( 𝑌 )

ii) 𝐸 ( 𝑌 / 𝑋=𝑥 )=𝑔 ( 𝑥 )=𝑥 + 1

2
(given)
𝐸 ( 𝐸 ( 𝑌 / 𝑋 =𝑥 ) ) =𝐸 (𝑌 )=𝐸 ( 𝑔 ( 𝑥 ) ) =𝐸 ( 𝑥 2 +1 )

𝐸 ( 𝑥2 +1 ) =𝐸 ( 𝑥2 ) +1
2
𝐸 ( 𝑥 ) =𝑉 ( 𝑥 )+ [ 𝐸 ( 𝑥 ) ]
2

𝑋 𝑁 (0 , 1)

=> 𝐸 ( 𝑋 ) =1 2

𝐸 ( 𝑥2 +1 ) =2=𝐸 ( 𝑌 )

NOW ,
 ( ( ))
𝑉𝐴𝑅 ( 𝑌 )=𝐸 𝑉𝐴𝑅
𝑌
𝑋
+❑❑ 𝑉𝐴𝑅 𝐸
𝑌
𝑋 ( ( ))
𝑉𝐴𝑅 ( 𝑌𝑋 )=𝑉𝐴𝑅 ( 𝑔 ( 𝑥 ) )=𝑉𝐴𝑅 (𝑥 +1) 2

𝑉𝐴𝑅 ( 𝑥 2+1 ) =𝑉𝐴𝑅 ( 𝑥2 ) + 0

2
𝑉𝐴𝑅 ( 𝑥 2) =𝐸 ( 𝑥 4 ) − ( 𝐸 ( 𝑥 2 ) )

Solving we get
𝑉𝐴𝑅 ( 𝑥 2) =2

𝑉𝐴𝑅 ( 𝑌 )=2+2=4

Hence
𝐸 ( 𝑌 )=2

𝑉𝐴𝑅 ( 𝑌 )=4
 Q11) 𝑋 ∼ 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 ( 𝑌 )
𝑌 ∼ 𝐿𝑜𝑔𝑛𝑜𝑟𝑚𝑎𝑙 ( 𝜇, 𝜎 2)

𝐸 [ 𝑋 ∣𝑌 =𝑦 )=𝑦

𝑉𝑎𝑟 ( 𝑋 ∣𝑌 = 𝑦 )=𝑦

𝐸 [ 𝑋 ]=𝐸 [ 𝐸 [ 𝑋 ∣𝑌 ) ]

𝐸 [ 𝑋 ]=𝐸 [ 𝑌 ]

𝑆𝑖𝑛𝑐𝑒 𝑌 ∼ 𝐿𝑜𝑔𝑛𝑜𝑟𝑚𝑎𝑙 ( 𝜇 , 𝜎 2 )
2
𝜎
𝜇+
𝐸 [ 𝑌 ]=𝑒 2
=𝐸[𝑋]

𝑉𝑎𝑟 ( 𝑋 )=𝐸 [ 𝑉𝑎𝑟 ( 𝑋 ∣𝑌 ) ] + 𝑉𝑎𝑟 ( 𝐸 [ 𝑋 ∣ 𝑌 ) )

𝐸 [ 𝑉𝑎𝑟 ( 𝑋 ∣𝑌 ) ] =𝐸 [ 𝑌 ]
2
𝜎
𝜇+
𝐸 [ 𝑉𝑎𝑟 ( 𝑋 ∣𝑌 ) ] =𝐸 [ 𝑌 ] =𝑒 2

𝑉𝑎𝑟 ( 𝐸 [ 𝑋 ∣𝑌 ) )=𝑉𝑎𝑟 ( 𝑌 )=( 𝑒 − 1 ) 𝑒

2 2
𝜎 2 𝜇+𝜎

2
𝜎
𝜇+
𝑉𝑎𝑟 ( 𝑋 )=𝐸 [ 𝑉𝑎𝑟 ( 𝑋 ∣𝑌 ) ] + 𝑉𝑎𝑟 ( 𝐸 [ 𝑋 ∣ 𝑌 ) ) =𝑒 + (𝑒 − 1 ) 𝑒
2 2
2 𝜎 2 𝜇+𝜎

The unconditional mean of X

2
𝜎
𝜇+
𝐸 [ 𝑋 ]=𝑒 2

The unconditional variance of X

2
𝜎
𝜇+
+ (𝑒 − 1 ) 𝑒
2 2
𝜎 2 𝜇+𝜎
𝑉𝑎𝑟 ( 𝑋 )=𝑒 2
 Q12) 𝑋 ∣𝑌 = 𝑦 ∼ 𝑁 ( 2 𝑦 , 𝑦 2)

𝑌 ∼ 𝑁 ( 200,100 )

𝑉𝑎𝑟 ( 𝑋 )=𝐸 [ 𝑉𝑎𝑟 ( 𝑋 ∣𝑌 ) ] + 𝑉𝑎𝑟 ( 𝐸 [ 𝑋 ∣ 𝑌 ) )

𝐸 [ 𝑋 ∣𝑌 =𝑦 )=2 𝑦
2
𝑉𝑎𝑟 ( 𝑋 ∣𝑌 = 𝑦 )=𝑦

𝐸 [ 𝑉𝑎𝑟 ( 𝑋 ∣𝑌 ) ] =𝐸 [ 𝑌 2 ]

𝐸 [ 𝑌 2 ] =𝑉𝑎𝑟 ( 𝑌 ) + ( 𝐸 [ 𝑌 ]) =100+2002=100+ 40000=40100

2

𝐸 [ 𝑉𝑎𝑟 ( 𝑋 ∣𝑌 ) ] =40100

NOW,
𝑉𝑎𝑟 ( 𝐸 [ 𝑋 ∣𝑌 ) )=𝑉𝑎𝑟 ( 2 𝑌 )

𝑉𝑎𝑟 ( 𝐸 [ 𝑋 ∣𝑌 ) )=400

𝑉𝑎𝑟 ( 𝑋 )=𝐸 [ 𝑉𝑎𝑟 ( 𝑋 ∣𝑌 ) ] + 𝑉𝑎𝑟 ( 𝐸 [ 𝑋 ∣ 𝑌 ) ) =40100+ 400=40500

The unconditional variance of x

𝑉𝑎𝑟 ( 𝑋 )=40500
Q13) Probability that the randomly selected person
does more than 2 hours of exercise
i) 𝑃 (𝑌 =2 )=0.25+0.25=0. .5
ii)For X and Y to be independent
𝑃 ( 𝑋 =𝑥 ,𝑌 = 𝑦 )=𝑃 ( 𝑋 =𝑥 ) 𝑃 ( 𝑌 =𝑦 )

Let , X=1 and Y = 2

𝑃 ( 𝑋 =1 , 𝑌 =2 )=0.05

𝑃 ( 𝑋 =1 ) =0.25

𝑃 (𝑌 =2 )=0.3

𝑃 ( 𝑋 =1 ) . 𝑃 ( 𝑌 =2 ) =0.25 .0.3=0.075 ≠ 𝑃 ( 𝑋 =1 ,𝑌 =2 )=0.05

𝑃 ( 𝑋 =1 ) . 𝑃 ( 𝑌 =2 ) ≠ 𝑃 ( 𝑋 =1 ,𝑌 =2 )

Hence X and Y ARE NOT INDEPENDENT

iii) 2
𝑅=( 3− 𝑌 ) ( 𝑋 +1 )
𝑃 ( 𝑋 =0 ,𝑌 =0 )=𝑅=9
𝑃 ( 𝑋 =0 ,𝑌 =1 )= 𝑅=4
𝑃 ( 𝑋 =0 ,𝑌 =2 ) =𝑅=1
𝑃 ( 𝑋 =1 , 𝑌 =0 )= 𝑅=18
𝑃 ( 𝑋 =1 , 𝑌 =2 )=𝑅=2

So, the possible values of R are

R=1,2,4,9,18
𝑃 ( 𝑅=1 )=𝑃 ( 𝑋 =0 , 𝑌 =2 )=0.25

𝑃 ( 𝑅=2 )= 𝑃 ( 𝑋 =1 ,𝑌 =2 )=0.05

𝑃 ( 𝑅=4 )= 𝑃 ( 𝑋 =0 , 𝑌 =1 )=0.3
 𝑃 ( 𝑅=9 )=𝑃 ( 𝑋 =0 , 𝑌 =0 )=0.2

𝑃 ( 𝑅=18 ) =𝑃 ( 𝑋 =1 ,𝑌 =0 )=0.1

IV) Expectation of R
❑
𝐸 ( 𝑅 )=𝑟 ∑ 𝑟 ⋅ 𝑃 ( 𝑅=𝑟 )

𝐸 ( 𝑅 )=1⋅0.25+ 2⋅0.05+ 4 ⋅ 0.3+9 ⋅0.2+18 ⋅0.1

𝐸 ( 𝑅 )=5.15
 Q14) 2
𝑓 𝑥𝑦 ( 𝑥 , 𝑦 )=𝑘 𝑥 𝑦 , 0< 𝑥< 𝑦 <1
2

i)To show k=18

❑ ❑

∫∫ 𝑓 𝑥𝑦 ( 𝑥 , 𝑦 ) 𝑑𝑥 𝑑𝑦=1 (Total probability)

❑ ❑

1 𝑦

∫∫ 𝑘 𝑥 2 𝑦 2 𝑑𝑥 𝑑𝑦=1
0 0

1 5
𝑦
∫𝑘 3
𝑑𝑦 =1
0

𝑘
=1
18
𝑘=18

ii) find 𝑓 𝑦( 𝑦)

Marginal density function of y is given by

❑
𝑓 𝑦 ( 𝑦 )=∫ ❑ 𝑓 𝑥𝑦 ( 𝑥 , 𝑦 ) 𝑑𝑥
𝑥

𝑦
𝑓 𝑦 ( 𝑦 )=∫ ❑ 18 𝑥 𝑦 𝑑𝑥
2 2

18 5 5
𝑓 𝑦 ( 𝑦 )= 𝑦 =6 𝑦
3

Hence, 𝑓 𝑦 ( 𝑦 )=6 𝑦
5

iii) 𝑃 ( 𝑌𝑋=0.75
> 0.5
)
0.75
𝑃 ( 𝑋 > 0.5
=∫ 𝑓 𝑥)
𝑌 =0.75 0.5 0.75
𝑥
0.75
𝑑𝑥 ( )
𝑓 (𝑥 ,𝑦 )
𝑓 𝑥
𝑦
( ) 𝑥
𝑦
= 𝑥𝑦
𝑓 𝑦 (𝑦)
 ( )
2 2
𝑥 18 𝑥 𝑦
𝑓 𝑥 =
𝑦 𝑦 6𝑦
5

0.75

( )
2
𝑋 > 0.5 3𝑥
𝑃 =∫ 𝑑𝑥
𝑌 =0.75 0.5 0.753

( )
3 3
𝑋 > 0.5 3 0.75 − 0.5
𝑃 = ⋅
𝑌 =0.75 0.75 3
3

Hence the required probability is 𝑃 ( 𝑌 =0.75 )=0.7037

𝑋 > 0.5