Scribd
Upload
10 views35 pages

Lesson 4 (Normal Distribution) Oct 2024

Uploaded by

2009 SAIFUL
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views35 pages

Lesson 4 (Normal Distribution) Oct 2024

Uploaded by

2009 SAIFUL
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 35

Normal Distributions

1
NORMAL DISTRIBUTION

Continuous distribution. It is also called the Gaussian


distribution. This is the most commonly used distribution
in statistics. Its graph is bell-shaped. Since it is a
continuous distribution, the total area under the
curve is one. The normal distribution has two
parameters (two numerical descriptive measures), the
mean 𝜇 and the variance 𝜎 2 :
X ~ N  ,( 2
)
2
PROBABILITY DENSITY FUNCTION
OF A NORMAL RANDOM VARIABLE :

The probability density function is rather complicated


function to evaluate. Do not memorize it. It is not
necessary

1 −( x −  )2 / 2 2
f (x) = e
 2

3
BELL SHAPED CURVE
The “Bell Shaped Curve” is a Normal Distribution. It
has:
• mean = median = mode
• symmetry about the center
• 50% of values less than mean and 50% greater than
the mean

4
MEAN AND VARIANCE OF NORMAL
DISTRIBUTION

X ~ N  ,( 2
)
X = 
X =
2 2

5
Normal Curve
For any normal population :
About 68% of the population is in the interval
μ ± σ.
About 95% of the population is in the interval
μ ± 2σ.
About 99.7% of the population is in the
interval μ ± 3σ.
6
99.7%

95%

68%

 − 3  − 2  − 1   + 1  + 2  + 3
7
THE STANDARDIZED OF NORMAL
DISTRIBUTION

Standard units tell how many standard deviations an


observation is from the population mean. The z-
score is an item sampled from a normal population
with mean 0 and standard deviation.
This normal population is called standard normal
population.
x−
z=
8

EXAMPLE 16

Assume that the heights in a population of women follow


the normal curve with mean μ = 64 inches and standard
deviation σ = 3 inches. The heights of two randomly chosen
women are 67 inches and 62 inches. Convert this heights
to standard units.

67 − 64
z= =1
3
62 − 64
z= = −0.67
3 9
Areas Under The Normal Curve

Z = - 0.67 Z=1
Z=0
10
EXAMPLE 2
Determine the area under the normal curve
to the left of z = 0.47.

From the z table, the area is 0.6808.


11
12
YOUR TURN

1. Determine the area under the normal


curve to the right of z = 1.38.
Solution : 0.0838

13
YOUR TURN

2. Determine the area under the normal


curve between z = 0.71 and z = 1.28.
Solution : 0.1386

14
EXAMPLE 3

Lifetimes of machines in a certain


application are normally distributed with
mean 65 hours and standard deviation 7
hours. Determine the probability that a
randomly chosen machine lasts between
55 and 70 hours.
15
(
X ~ N 65 , 7 2
)
P (55  x  70 )
55 − 65 70 − 65
z= = −1.43 z= = 0.71
7 7

16
P (55  x  70 )
= P (− 1.43  z  0.71)

A2 A1

Z = - 1.43 Z = 0.71
Z=0

17
A1 = 0.2611
A 2 = 0.4236
P (55  X  70) = P ( −1.43  Z  0.71)
= P ( 0  Z  0.71) + P ( 0  Z  1.43 )
= A1 + A2
= 0.2611 + 0.4236
= 0.6847

18
THE CENTRAL LIMIT THEOREM

 The most important result in statistics.


 The theorem says :
‘if we draw a large enough sample from a
population, then the distribution of the sample
mean is approximately normal, no matter what
population it was drawn from.’
 How large is large enough?
 Empirical evidence suggests that :
A sample size of > 30. 19
Let X1,…,Xn be a simple random sample from a population
with mean μ and variance σ2.

X 1 + ... + X n
Let X = be the sample mean.
n

Let S n = X 1 + ... + X n be the sum of the sample


observations. Then if n is sufficiently large,

  2

X ~ N   , 
 n 
20
EXAMPLE 4

Let X denote the number of flaws in 1 inch length of


copper wire. The probability mass function of X is
presented in the table.
One hundred wires are sampled from this population. What
is the probability that the average number of flaws per wire
in this sample is less than 0.5?
x P(X=x)
0 0.48
1 0.39
2 0.12
3 0.01
21
Population mean, μ = 0.66
Population variance, σ2 = 0.5244
Let X1,…,X100 denote the number of flaws in the 100 wires.
Need to find P ( X  0 .5 )
N = 100 (large sample - follows Central Limit Theorem)

X ~ N (0 .6 6 , 0 .0 0 5 2 4 4 )
z-score : 0 . 5 − 0 . 66
z= = − 2 . 21
0 . 005244
22
P ( X  0 . 5 ) = P (z  − 2 . 21)
= 0 . 0136

0.0136

-2.21

Only 1.36% of samples of size 100 will have


fewer than 0.5 flaws per wire.
23
Approximation of Binomial to
Normal Distributions

24
When the number of trials becomes large,
evaluating the binomial probability function
by hand or with the calculator is difficult.
Thus, a binomial distribution with parameters
n and p can be approximated by a normal
distribution if

np  5 and nq  5
25
APPROXIMATION OF BINOMIAL TO
NORMAL DISTRIBUTIONS

X ~ Bin (n, p ) → X ~ N (np, npq )

when

n  30 and p  0.5

26
The Continuity Correction
Continuity Correction is needed because
we are approximating a discrete probability
distribution with a continuous distribution.

27
THE CONTINUITY CORRECTION

For example: to compute the probability of


getting exactly 10 heads, determine
the probability of getting between 9.5
and 10.5 heads.

28
9.5 - 10.5
The Continuity Correction

We are using the area under the


curve to approximate the area of the
rectangle.

29
9.5 - 10.5
The Continuity Correction

In general, if a and b are two integers,


a ) P ( X  a ) = P ( X  a − 0.5 )
b ) P ( X  a ) = P ( X  a + 0.5 )
c ) P ( X  a ) = P ( X  a + 0.5 )
d ) P ( X  a ) = P ( X  a − 0.5 )
e ) P ( a  X  b ) = P ( a − 0.5  X  b + 0.5 )
f ) P ( a  X  b ) = P ( a + 0.5  X  b − 0.5 )
g ) P ( X = a ) = P ( a − 0.5  X  a + 0.5 ) 30
EXAMPLE 5MPLE 10

A regular tetrahedral shaped dice with its faces


labelled 1, 2, 3 and 4 is tossed 200 times. Determine
the probability of obtaining
a) more than 60 times the digit 4
b) at most 50 times the digit 4
c) exactly 50 times the digit 4

31
Let X be the random variable which represents the
number of times the digit 4 appears (success)
 1
X ~ Bin 200 , 
 4
Since n is large, np = (200)(0.25 )
= 50
np  5
nq = (200 )(0.75 )
= 150
nq  5
Both are greater than five. We use normal approximation
32
 = np = 50
 2 = npq = (50 )(0.75 ) = 37.5

Therefore, X ~ N (50 , 37.5 )


a ) P ( X  60 ) → P ( X  60.5 ) continuity correction
 60.5 − 50 
= P Z  
 37.5 
= P ( Z  1.715 )
= 0.5 − P ( Z  1.715 )
= 0.5 − 0.4568
= 0.0432
33
b ) P ( X  50 ) → P ( X  50.5 )
 50.5 − 50 
= P  Z  
 37.5 
= P (Z  0.0816)
= 0.5327
c ) P ( X = 50 ) → P (49.5  X  50.5 )
 49.5 − 50 50.5 − 50 
= P  Z 
 37.5 37.5 
= P (− 0.0816  Z  0.0816)
= 2(0.5327) − 1
= 0.0654
34
YOUR TURN

A radar unit is used to measure speeds of cars on a


motorway. The speeds are normally distributed with a mean
of 90 km/hr and a standard deviation of 10 km/hr. What is the
probability that a car picked at random is travelling at more
than 100 km/hr?

35

Ans: 0.1587

You might also like