Sie sind auf Seite 1von 13

The Use Of Combining Two Linear Independent Normal

If X and Y are independent normal random variables


2
2

so X ~ N ( 1 1 ) and Y ~ N ( 2 2 ) .
If a and b are constants,

E (aX+ bY)

= aE(X) + bE(Y)
= a 1 + b 2

Var (aX + bY)

= a Var (X) + b Var (Y)


= a 1 + b 22
2

then

aX + bY ~ N (a1+ b 2 , a 1 + b 22 )
2

aX bY ~ N (a1 - b 2 , a 1 + b 22 )
2

When

a=b=1

then,

X + Y ~ N ( 1+ 2 , 12+ 22)
X Y ~ N ( 1 - 2 , 12 + 22)

Example
If T~ N (, 25) and P( T > 27.5 ) = 0.6915, find ,
Solution

= 25 , = 5
T - 25.7 -
P(

) 0.6915
5
5
Since the value of the probability is bigger than 0.5 then

P(Z

25.7 -
) 0.6915 must be negative.
5

T-
25.7 -
k then P(Z
Let
) 0.6915
5
5
25.7 -
Q(
) 0.6915
5

If

Q(k) = 1-Q(-k)
0.6915 = 1-Q(-k)
Q(-k) = 1-0.6915
Q(-k) = 0.3085
-k = 0.5
k = -0.5

then 25.7 - 0.5


5
27.5- = -2.5
= 27.5 + 2.5
= 30

Example
The marks in an examination were normally distributed with
mean and variance . 10 % of the candidates had more than 75
marks and 20% had less than 40 marks. Find the values of and .

Solution
P ( X > 75 ) = 0.1
P ( X < 40 ) = 0.2

X 75
P

0.1


75

P Z
0.1

75
Q
0.1

Q(1.281)=0.1

Then Q 75 1.281

75 - = 1.281 - - - - - - - - - - - - - - - - - (1)
X 40
P

0.2


40

P Z
0.2

40

1- P Z
0.2

40

P Z
1 0.2

40

P Z
0.8

40 -
Let

Since the value of the probability is bigger than 0.5 then P Z

40
0 .8
be negative, then Q

0.8 = 1 - Q(-k)
Q(-k) = 1 - 0.8
Q(-k) = 0.2
From tables, Q(0.841) = 0.2
-k = 0.841
k = -0.841
then,

40 -

0.841

40 - = -0.841 - - - - - - - - - - - - - - - - - - - - - (2)

40
must

equ (1) equ (2)

35 =1.281 + 0.841
35 =2.122
=16.49

substitute into equation (2)

40 - = -0.841(16.49)
40 - = -13.87
= 40+13.87
= 53.87

Example
If X ~ N (80 , 36) and Y ~ N (90 , 484), find
(a) P( X + Y >160)
(b) P(Y X < 0)
(c) P(150 < X + Y < 210)
(d) P(3X > 2Y)

Solution
(a) X + Y ~ N( 80 +90, 36 + 484 )
X + Y ~ N( 170, 520)

Let
Then

X+Y=T
P( X Y 160) P(T 160)

T 170 160 170


P

520
520

= P(Z>-0.4385)
=1 - P(Z>0.4385)
=1 - 0.3333
=0.6777

(b) Y X ~ N ( 90 80 , 36 + 484 )
Y X ~ N (10 , 520 )
Let Y X = T
Then
P( Y - X < 0 ) = (T < 0 )
T 10 0 10
P

520
520
= P(Z< -0.4385)
= P(Z> 0.4385)
= 0.3333

(c) X + Y ~ N ( 170 , 520 )


Let X + Y = T
Then
P( 150 < X + Y < 210 ) = P(150 < T < 210)
210 170
150 170
P
Z

520
520

= P(-0.877 < Z < 1.754)


= 1- P(Z > 0.877) P(Z > 1.754)
= 0.7701

(d) P(3X > 2Y) = P(3X-2Y>0)


3X-2Y ~ N ( 3(180)-2(90) , 3(36)+2(484) )
~ N ( 240 180, 324 + 1936 )
~ N ( 60, 2260 )
Let
Then

3X 2Y = T
P( 3X - 2Y > 0 ) = P ( T > 0 )
T 60 0 60
P

2260
2260

= P ( Z > -1.2621 )
= 1 - P ( Z > 1.2621 )
= 1 0.1034
= 0.8966

Example
The height of women in a town under normal distribution is with a mean of
165 cm and a variance of 6 cm. The height of men, normally distributed is
with a mean of 174 cm and variance 7 cm. If a woman and a man are
chosen randomly, find the probability that :
(a) A woman is taller than a man
(b) A man is 5 cm taller than a woman

Solution
W ~ N ( 165 , 62 )
M ~ N ( 174 , 72 )

(a)

P ( W > M ) = P (W M > 0)
W M ~ N ( 165 -174 , 62 + 72 )
~ N ( -9 , 85)

Let

WM=T

Then

P(WM>0) =P(T>0)
T 9 09
P

85
85
= P ( Z > 0.9762 )
= 0.1645

(b)

P(M>W+5)=P(MW>5)
M W ~ N ( 174 165 , 62 + 72 )
~ N ( 9 , 85 )

Let
Then

MW =T
= P ( M W > 5)
= P ( T> 5)
T 9 59
P

85
85

= P ( Z > 0.4339 )
= 1- 0.3322
= 0.6678

Das könnte Ihnen auch gefallen