Normal probability distribution

Friday, February 12, 2021 19:44

Some examples of variables associated with natural phenomena that follow the normal model are:
• Morphological characteristics of individuals such as height;
• Physiological characteristics such as the effect of a drug;
• Psychological characteristics such as IQ;
• Noise level in telecommunications;
• Errors made when measuring certain magnitudes

Given a continuous random variable X , it is called normal if its density function is defined by

1
f(x) = _____ e 200/ x € R
V2n0

Unit 04 page 1
To find a probability we have

b b
P\a < X < b] = í f(x)dx = Í ———e-2a ) dx
J. J \2n(j
to to

Which is interpreted as follows

CUMULATIVE DISTRIBUTION FUNCTION

We define

F(K)=pX<xj= f(dt= -1e-2(F)

JJ V2n0

Which is interpreted graphically as

Then we obtain that

P\a < X < b] = P[X < b] - P[X < a]

Since it can be interpreted as a difference in areas

Unit 04 page 2
Example: If X has a normal distribution with
u =60 o = 8 . Find the following probabilities

a) P[X<53]=0.1908

b) P[X< 62] =?

pXc] 0.5989
P[X

>55]=1-P[X<=55]
u =60 o= 8
=AD 2.6

- 0 9340

c) P[X >55]=?

Unit 04 page 3
d) P[45 < X < 61] =?
P[45 < X < 61] = P[X < 61] - P[X < 45]
= 0.5497 - 0.0304
= 0.5193

P[a < X < b] = P[X < b] - P[X < a]

u =60 a= 8

e) P[X>21]=1-P[X<=21]=0.99999964 f) P[40 < X

<55]=P[X<55]-P[X<=40]
=0.2660-0.0062
=0.2598

40 55=60

Example:
A soft drink dispensing machine is calibrated to
dispense 250 milliliters of soft drink per glass. The
standard deviation is 4 milliliters. What is the probability that a machine dispenses:

a. Between 240 and 255 milliliters of soda?

b. 260 milliliters or more?
c. Between 270 and 280 milliliters of soda?
d. How much soda is dispensed in the largest 2% of soft drinks?

X: amount of soda that the machine fills in the glass

P[240<X<255)
PC X>-260]
PL20<x <zg ^

Solution:

Unit 04 page 4
 X: Amount of soft drink that is filled in a glass (milliliters)

a) P[240<X<255]=P[X<255]-P[X<240] =0.8944-0.0062 =0.8881

The probability that the machine dispenses between 240 and 255 milliliters of soda is 88.81% b) P[X>=260]=1-
P[X<260]=0.0062

The probability that the machine dispenses 260 milliliters of soda or more is 0.62% c)
P[270<X<280]=P[X<280]-P[X<270]
= 1-0.999999713
=2.86652E-07=2.86652 x 10-7
=0.000000286652

The probability that the machine dispenses between 270 and 280 milliliters of soda is 0%

d) How much soft drink is dispensed in the largest 2% of soft drinks?

Unit 04 page 5
 PLx>x)= or 02
PCx4x]=0.98
x= 258.21
It is dispensed from 258.21 milliliters or more in the largest 2% of soft drinks

x=? 7%

superior

either
p[X <X]=0.93

x=? 4%

lower

x=?
X=242 92
PUX<x]=0.04

Example:
In a company, its workers use the computer an average of 27
hours per week. Assume that the usage time corresponds to a
normal distribution and that the standard deviation is 8 hours.
a. What is the probability that a randomly chosen worker uses
the computer for less than 11 hours?
b. What percentage of workers use the computer for more
than 40 hours?
c. A worker is classified as an extraordinary user if he or she is
in the top 20% in terms of hours of use. How many hours
=27 must a worker
use the computer
to be considered
0=g an extraordinary
user?

X~ N(j,o:)
a) PLX<8]=?

Ha) P[X>40=?

Unit 04 page 6
 c) P[X,7=02
Solution:
H=23

0=g
X* Daily hours of the computer

a) P[x<n] = 0 022”

b) P[x > 40] - 1 ■ P[x640]

=I -09499
= 00521

c) P[X>]=0.2

PLx<x]=0.?

% =3333

Example: The amounts of money requested in mortgage loan

applications at Dawn River Federal Savings have an approximately
normal distribution with a mean of $70,000 and a standard deviation
of $20,000. One morning you receive a credit application. What is
the probability that:

a) Is the requested amount $80,000 or more?

b) Is the requested amount between $65,000 and $80,000?
c) Is the requested amount no more than $65,000?
d) 20% of the credits are greater than what amount

J = ?o ( ooo
or — 20,000

X: amount of money requested in a credit application

o) P[X>80.0003=?
b) P [ 65,000 <X= g0,000] - ?

c PC X <65000] =?

d) PE X>x,0.2

Example: According to the Bureau of Labor Statistics, the average

monthly pay for a worker in the United States was $441.84. Suppose
that the available data indicate that wages are normally distributed with
a standard deviation of $90.

a) What is the probability that a worker earns between $400 and

$500?

Unit 04 page 7
 b) How much does a worker have to produce to be in the top
20% of those who earn a salary?
c) For a worker chosen at random, what is the probability that he
earns less than $450?

Unit 04 page 8
 STANDARD NORMAL DISTRIBUTION

If Z is a random variable that has a normal distribution with

u
=0 and variance o 2 =1 then Z is called standard normal
distribution, its density function

1 _22
f(z)= —=e 2
V2n

F(z) = P[Z < z] = J f(t)dt = J Tte 2 dt

In general we have to perform a transformation by changing

the variable

1 -1/3-2
f(x) = ______ e Z00/ x€RN(u,02)
V2T0

F(x) = px <xj= f f(t)dt = í ______________e 2( 6 ) dt J J

V2T0

Unit 04 page 9
 Example:

Using the standard normal distribution table, find

a) P[Z<2.35]=? P[Z<2.35]=0.9906

2.35 — 2.3-0.05

2.35 =2.3 +0.05

b) P[Z<3]=0.998650102

Unit 04 page 10
c) P[Z> 0.42]=1- P[Z<=0.42]=1- 0.662757273=0.33727272

2=042
z = 04-0 02

d) P[Z>-2.15]=1-P[Z<= -2.15]=1- 0.015777607=0.984242

e) P[-1.23 < Z < 3.21] = P[Z < 3.21] -P[Z < -1.23]
= 0.9993 - 0.1093
=0.89

Example:

If X has a normal distribution with

• u =60 a = 8 . Find the following probabilities

a) P[X < 53] = 0.19078695

P[X < 53] = PÍ— < 53.60 = P[Z < -0.875] and P[Z < -0.88] = 0.1894 LO 8 J

53 - 60
-0.875

Unit 04 page 11
b) P[X > 21] =1-P[X< 21]
_ 1 _ p xu < 21-60
—
•Lo3 8
= 1 -P[Z < —4.875
= 1 -P[Z < -4.88]
= 1-0
P[X > 21] = 1
-4.88

P[Z < -4] = 3.16712E - 05 = 0.0000316712 s 0

c) P[40 < X <55]=???

P[40 < X <55]=P[X<55]-P[X<=40]

=0.2660-0.0062
=0.2598 40 55 CO
2"

P[40 < X <55]=P[X<55]-P[X<=40]

- xH 40—60 ,
8 ]-P[ ]

=P[ Z < -0.62]-P[ Z < -2.5]

= 0.2676 -0.0062
=0.2614
Z,=-2.5
62

Unit 04 page 12
Example: If X has a normal distribution with
• u =20 a = 3 . Find the following probabilities

a) P[X<14]
2 = X-
either
£ at 14 -20
3

Unit 04 page 13
 b) P[X>22] = < -Px$2]
=i - P[*-M,22,20]

=4 -P[2<o.61]

-, - 0.74z4

= 0.2544

c) P[6 < X <23] = P[X<23 -P[<6

PL]
z 1 (45

Example: Inverse Probability

Unit 04 page 14
PL-1 <£<•]- ? r -
p[- 1<2,0=PC2<n- P<-'
— 0.8413 - 0.1586

0.6827

EXAMPLE: Find z if P[Z< z]=0.86 is verified

x EXAMPLE: Find if P[X< x]=0.35 is verified with mean 50 and standard deviation 7

Unit 04 page 15
 PEX<xJ = O 35
PL:]

=-039

=-0.39

X-50 _ -0.39 Y

X £4121

P[X¿ 4129] £035

Unit 04 page 16
 N(o,4)

P[-1 < Z < 1] = P[Z < 1] - P[Z < -1]

0.841344746 - 0.158655254
= 0.682689492 ~ 68.27%

P[—2 < Z < 2] = P[Z < 2] - P[Z < —2]

0.977249868 - 0.022750132 = 0.954499736 ~ 95.45 %

Unit 04 page 17
 P[-3 < Z < 3] = P[Z < 3] - P[Z < -3]

= 0.998650102 - 0.001349898
= 0.997300204 ~99.73 %

Example: A statistical analysis of 1000 long-distance telephone calls made from a telephone exchange indicates that the duration
of these calls has a normal distribution with a mean of 129.5 seconds and a standard deviation of 30 seconds.

a) What is the probability that a particular call lasted between 89.5 and 169.5 seconds?
b) How many calls lasted less than 60 seconds or more than 150 seconds?
c) What should be the duration of a particular call if only 1% of all calls are shorter?

u= 427.5
O'-30
X: Time one lasts. Wamd telenica •
long. distance (Sec)

a) P[89.s 4 Xe 169.5J = ?

PE875 X 169.5]= POX<161.5-P[X $895

= 0.90918878 - O. 9121422

-0 g4957956 ~ 84,367

b) PLX<60V X> (533 “ PLx< 60+ PLx>IS)

= P[x¿60] 4-PX,4S03

Unit 04 page 18
 — 0.04026 t I -01528
— or 2544

For n -1OOD Ua.YYWi.daLS Teleg

or 2546 (100)= 25146

C) P[X<x]=0.01

x=???

P[Z<z]=P[X<x]=0.01

P[Z<z]=0.01

z=?? z = -2.33

x—u
z =----------
EITHER

x = 59.6

Unit 04 page 19