STAT 400 Stepanov

UIUC Examples for 3.3 Dalpiaz

1. At Initech , the salaries of the employees are normally distributed with

mean µ = $36,000 and standard deviation σ = $5,000.
a) Mr. Smith is paid $42,000. What proportion of the employees of Initech
are paid less that Mr. Smith?

 42,000 − 36,000 
P( X < 42,000 ) = P Z <  = P( Z < 1.20 ) = 0.8849.
 5,000 

b) What proportion of the employees have their salaries over $40,000?

 40,000 − 36,000 
P( X > 40,000 ) = P Z >  = P( Z > 0.80 ) = 1 – 0.7881 = 0.2119.
 5,000 

c) Suppose 10 Initech employees are randomly and independently

selected. What is the probability that 3 of them have their salaries
over $40,000?

Let Y = number of employees (out of 10) who have salaries over $40,000.
Then Y has Binomial distribution, n = 10, p = 0.2119 ( see (b) ).

P( Y = 3 ) = 10 C 3 ⋅ (0.2119 ) 3 ⋅ (0.7881) 7 = 0.2156.

d) What proportion of the employees have their salaries between $30,000

and $40,000?

 30,000 − 36,000 40,000 − 36,000 

P( 30,000 < X < 40,000 ) = P <Z< 
 5,000 5,000 
= P( – 1.2 < Z < 0.80 ) = 0.7881 – 0.1151 = 0.6730.
e) Mrs. Jones claims that her salary is high enough to just put her among the
highest paid 15% of all employees working at Initech . Find her salary.

Need x = ? such that P( X > x ) = 0.15. ( area to the right is 0.15 )

First, need z = ? such that P( Z > z ) = 0.15.
z = 1.04.

X = µ + σ Z. x = 36,000 + 5,000 × 1.04 = $41,200.

f) Ms. Green claims that her salary is so low that 90% of the employees
make more than she does. Find her salary.

Need x = ? such that P( X > x ) = 0.90. ( area to the right is 0.90 )

First, need z = ? such that P( Z > z ) = 0.90.
z = – 1.28.

X = µ + σ Z. x = 36,000 + 5,000 × ( – 1.28 ) = $29,600.

2. Suppose that the lifetime of Outlast batteries is normally distributed

with mean µ = 240 hours and unknown standard deviation. Suppose
also that 20% of the batteries last less than 219 hours. Find the standard
deviation of the distribution of the lifetimes.

Need σ = ? Know P( X < 219 ) = 0.20.

First, need z = ? such that P( Z < z ) = 0.20.

z = – 0.84.
X = µ + σ Z. 219 = 240 + σ × ( – 0.84 ).
– 21 = σ × ( – 0.84 ).
σ = 25 hours.
 Let X be normally distributed with mean µ and standard deviation σ.
Then the moment-generating function of X is

e μt+σ
2 2
t 2
MX(t ) = .

∞
e − ( x −μ ) 2 σ2
1 2
MX( t ) = E( etX ) = ∫ et x ⋅
2π σ
dx
−∞

∞ ∞
e t (μ + σ z) ⋅ 1 dz = e μ t + σ t 2 ⋅
1
e −( z −σ t )
2 2 2
−z 2 2
∫ ∫
2
= e dz
2π 2π
−∞ −∞

e μt + σ
2 2
= t 2,

1
e −( z −σ t )
2
2
since is the probability density function
2π
of a N ( σ t , 1 ) random variable.

Let Y = a X + b. Then M Y ( t ) = e b t M X ( a t ).

Therefore, Y is normally distributed with mean a µ + b and variance a 2 σ 2

( standard deviation | a | σ ).

1. (continued)
g) All Initech employees receive a memo instructing them to put away
4% of their salaries plus $100 per month ( $1,200 per year ) in a special
savings account to supplement Social Security. What proportion of the
employees would put away more than $3,000 per year?

Y = 0.04 X + 1,200. P( Y > 3,000 ) = ?

Y > 3,000 ⇔ X > 45,000.

 45,000 − 36,000 
P( X > 45,000 ) = P Z >  = P( Z > 1.80 ) = 1 – 0.9641 = 0.0359.
 5,000 
 OR

µ Y = 0.04 × 36,000 + 1,200 = $2,640, σ Y = 0.04 × 5,000 = $200.

 3,000 − 2,640 
P( Y > 3,000 ) = P Z >  = P( Z > 1.80 ) = 1 – 0.9641 = 0.0359.
 200 

3. Suppose the average daily temperature [in degrees Fahrenheit] in June

in Anytown is a random variable T with mean µ T = 85 and standard
deviation σ T = 7. The daily air conditioning cost Q , in dollars, for
Anytown State University, is related to T by

Q = 120 T + 750.
Suppose that T is a normal random variable. Compute the probability
that the daily air conditioning cost on a typical June day for the university
will exceed $12,210.

Q has Normal distribution.

µ Q = 120 µ T + 750 = 120 ⋅ 85 + 750 = $10,950.

σ Q2 = ( 120) 2 ⋅ σ T2 = (120) 2 ⋅ 7 2 = 840 2. σ Q = $840.

 12 ,210 − 10 ,950 
P( Q > 12,210 ) = P Z >  = P( Z > 1.50 ) = 1 – Φ ( 1.50 ) = 1 – 0.9332 = 0.0668.
 840 

OR

12,210 = 120 T + 750. ⇔ T = 95.5.

P( Q > 12,210 ) = P( T > 95.5 )

 95.5 − 85 
= P Z > 
 7 
= P( Z > 1.50 )
= 1 – Φ ( 1.50 )
= 1 – 0.9332
= 0.0668.
4. Consider a random variable X with the moment-generating function
2
MX(t) = e 3 t + 8 t = exp ( 3 t + 8 t 2 ).

a) Find P ( X > 0 ).

e μt + σ
2 2
Normal distribution: MX(t ) = t 2
.

X has a Normal distribution with µ = 3 and σ 2 = 8.

2
⇒ /
⇒ E ( X ) = µ = 3, Var ( X ) = σ 2 = 16. σ = 4.

 0−3
P( X > 0 ) = P Z >  = P ( Z > – 0.75 ) = 1 – Φ ( – 0.75 ) = 1 – 0.2266 = 0.7734.
 4 

a) Find P ( – 1 < X < 9 ).

 −1− 3 9−3
P ( – 1 < X < 9 ) = P <Z<  = P ( – 1.00 < Z < 1.50 )
 4 4 

= Φ ( 1.50 ) – Φ ( – 1.00 ) = 0.9332 – 0.1587 = 0.7745.

___________________________________________________________________________

EXCEL:

= NORMSDIST( z ) gives P( Z ≤ z ) = Φ( z )

= NORMSINV( p ) gives z such that P( Z ≤ z ) = p

= NORMDIST( x , µ , σ , 1 ) gives P( X ≤ x ), where X is N( µ , σ 2 )

= NORMDIST( x , µ , σ , 0 ) gives f ( x ), p.d.f. of N( µ , σ 2 )

= NORMINV( p , µ , σ ) gives x such that P( X ≤ x ) = p,

where X is N( µ , σ 2 )
5.* Show that the odd moments of N ( 0, σ 2 ) are zero and the even moments are

μ 2n =
( 2 n )! σ 2 n
2 n ( n )!

Taylor Formula:

∞ tr ∞ tr ∞ tr
MX(t) = ∑ M (r ) (0 ) =  X r  =
∑ r!  
E  ∑ r! μ r .
r =0 r ! r =0 r =0

Since X is N ( 0, σ 2 ),

 σ 2 t 2  ∞ σ 2 n t 2n
M X ( t ) = exp   = ∑
 2  n
n =0 2 n!

Therefore,

if r is odd, μ r = 0,

if r = 2 n is even,
σ 2n
=
1
μr ⇒ μ 2n =
( 2 n )! σ 2 n .
2 n n! r ! 2 n ( n )!

OR

∞
Def Γ( x ) = ∫u
x −1 e − u du , x>0
0

Γ( x ) = (x – 1 ) Γ(x – 1)

Γ ( n ) = ( n − 1 )! if n is an integer

1
Γ  = π
2
 ∞ 2 2σ 2 ∞ 2 2σ 2
e−x e−x
1 2
μ 2n = ∫ x 2n dx = ∫ x 2 n dx = …
−∞ 2π σ 0 2π σ

x2 x dx du σ
u= du = dx =
2σ2 σ2 2u

∞ −u
1  1
u n −1 2 e
1
∫2 σ du = 2 n σ 2 n
… = n 2n Γn +  .
0 π π  2

 1  1  3  5 3 1
Γ  n +  =  n −  ⋅  n −  ⋅  n −  ⋅ ... ⋅ ⋅ ⋅ π
 2  2  2  2 2 2

⋅ ( 2n − 1 ) ⋅ ( 2n − 3 ) ⋅ ( 2n − 5 ) ⋅ ... ⋅ 3 ⋅1 ⋅ π
1
=
2n

=
1
⋅
( 2 n )! ⋅ π
2 n ( 2n ) ⋅ ( 2n − 2 ) ⋅ ( 2n − 4 ) ⋅ ... ⋅ 4 ⋅ 2
=
1 ( 2 n )! ⋅ π
⋅ n
2 n 2 ⋅ ( n )!

⇒ μ 2n =
( 2 n )! σ 2 n
.
2 n ( n )!