Chap ter-10

-
----S peci al Proba bility Distri butio ns

. . In th is chapter we shall discuss four spe-cial types of probability

dtstnbution ·
. . . s are based on discrete
'-"·· · 'l'he three d1stnbution rand om
.
v:iri~bles, so these are called discrete probability distribut:0 ns and one
di stnbution is basc<l a 11 ,~ontinuour. random variable so. it is called
continuou s probability distribution.
All thesP- special types of probability distributio ns are explained one
by one in the following sections,

Special Discrete Probability Distributions

Following are the three special types of discrete probability
distributio ns:
(i) Binomial Probabilit y Distribution
(ii) Hypergeometric Probabilit y Distribution
(iii) Poisson Probabilit y Distribution

Blnomlal Probability Distribution

Binomial Probabilit y Distribution is a mathemat ical formula used to
calculate the probabiliti es of the possible values of the random variable.
The possible values of X are 0, 1, 2, ............. n. Hence, the Binomial
Probabilit y Distributio n of the random variable xis given by the following
formula

P(x ) = ncx P X
•q 11-x . Ifor x -_0, 1, 2, .............n

337.
SSS Statisdcs Problems & Pr'!¢ce

where n = No. of independent trials

x = No. of successes in n trials
p = Probability of a success in a single trial remains constant
from trial to trial
q === 1- p = probability of a failure in a single trial

Remarks:
· (i) Binomial Distribution has two parameters i.e. n and P·

(ii) 1'1ean = np and Variance = npq

(iii) Binomial distribution will also be applic~ble when '1 individuals
or objects are drawn with replacement.

Example 1
A fair coin is tossed 4 times. What is the probability of obtaining (a) 3 heads
(b) 2 heads (c) 1 head (d) at least 2 heads (e) no head
Solution:
(i) n = 4 = No. of tosses I No. of trails (i.e independent trails)

(ii) x = No. of heads/ No. of successes

(iii) each toss of a coin ( i.e trial) has two possible outcomes i.e. head ( success)
and tail (failure)
1
p = P( a Head) = P(Success) = - remains same from trial to trial and
2

q = P(a Tai 1) 1 - p = 1 - !2 = !2
Therefore the random variable X follows the Binomial Distribution with
1
p =- and n = 4
2
HenceP(x) = ncX PX .q"-x - "t'.
ACX ( -~ )x •(-!)• ~
Now find probabilities

. (a) P( 3 -heads)= P(x = 3) = 4c3 GJ' {½J

 S ecial Probabili Distributiot1s
, . #JO -------339
Cp
(b)P(?-heads)·P(x=2)=4c2 liJHf I 2
=

(C) P(l -hea~) · P(x = I) C, \ (i)' Hf' ! =

(d) P( at least 2 hea<ls) = P(x = 2) +P(x = 3) + P(x = 4)

(HHf (D'Hf'
2
4 4 4
= c2 + c3 + C4GJHf'

P(at least 2 heads) =I + ! + _!_ = .!_!

Hr
8 4 16 16

co(H
O
4
(e)P(nohead)=P(x = 0) = _ -1~

Example 2
The probability that a patient recovers from a delicate heart operation is 0.9. What is
the probability that exactly 5 of the next 7 patients having this operation survive.
Solution:
n =No.of patients having operation= 7
x =No.of patients survive/ No. of successes= 5
p = P(success) = P( patient survive)= 0.9
q = 1 - 0.9 = 0.1
then P(x = 5) = 7c5 (09) 1 • (01) ?- 5 = (21)(059049)(0.01) = 0124

Examp1e3
The probability that a student is not a smoker is 2/3. Find the probability that out
of 5 students selected at random (i) 4 are smokers (ii) at least 4 are smokers.
Solution:

n = No. of students=5
x =No.of smokers / No. successes
2 1
P == P( success)= P( smoker) = 1 - =
3 3
340 - - - - - - - - - - - : ; - -_ _ _ _ __.,!11~7!!'!!!'=!!."'~!!!'!!'~!.!!~
q = P( failure)= P( not smoker) - ~3

c4 (31)" . (2)'
3-
4
5
(i) TbenP(x = 4) = = ;:3

(ii) P( at least 4 are smokers)• P(x > 4) • P{ x • 4) + P ( x = 5)

5c4 (-3t ) . -(2)3

4 5
-
4
+5c )s . -3
( t-
5 3
(2)' -' =24311-
Example4
Compute the mean and variance of the Binomial distribution when
n = 10 and p = 0.6.
Solution:
Mean= np = (10) ( 0.6) = 6 and
Variance = npq where q = 1 - p = 1 - 0.6 == 0.4
Variance = npq = (10) (0.6) (0.4) =2..t
Examp1e5

Mean ofa binomial distribution is

Solution:
Mean=np
then np = 20 ............... .
and variance = npq
then npq = 16 ............... .
Putting (1) in (2)

then 20q = 16 and q = 16 =

20

Since p + q = 1 then p = 1

As np = 20 and p = ! ...,_.,._.
5
 . I Probability Dbilrlbutlmts • 341
r1, # JO Speera
~- ergeometric ,ro bab lllt y Distribution
ffyp
If there is a pop ulat ion of N item s, classified as;

Type-A Type-B Total items

No. of successes No. of failures

1--
- - N
k N- k
~~
at random without repl ac~ ~en t,
Suppose a ran dom sam ple of n-items is selected
in a random sample of size n is
e rob abil ity of 'x' successes. and (n- x) failures
y

P(x ) - for x = 0, 1, 2, ............ ,n

all pos sibl e valu es of the random

The above formula give s the probabilities of
ility Distribution.
variable x and is call ed Hyp erge ome tric Pro bab
Exa mp le 1
from 3 wom en and 5 men. Fin d the
A committee of size 5 is to be sele cted at rand om
en on the committee.
probability dist ribu tion for the num ber of wom
Sol utio n:

Population of size N= 8 is clas sifi ed as

I
I
Women
k= 3
/ Men
/N -k= 8-3 =5 /

Let X = No. of wom en on the com mit tee/ No.

/
N= 8
Total

of suc ces ses, then the pos sibl e

l
values of X are 0, 1, 2 and 3.
•
Therefore ..

( 3c1 )( sc4) -15 -

(3co )( scs) =-1 , P(x - 1) - - 56
P(x - 0) - Be
Be 56 5
5
 Statistics Problems & Practice
342 - -- --
I ·\ '., ' ' ) C )
\. ... \ J 30
(3c3) (5c2) -=
10
= and P(x = 1) -=
P( X = ..: ) ~
8 56 Sc 56
cs 5

The distribution may be written in tabular fonn as

-~-
~=r=---()
"":.
;,=-~
X
1 2 3
II 1 15 30 10
P(X= x) - - - -
!I
56 56 56 56
t"'· ~= --

The distribution may also be written in the· form of formula as

for X = 0, 1, 2 , 3

of the random
This fonnula is called hypergeometric . probability distribution
variable X, where X== No. of women on the committee.
Example 2

A box of 8 screws contains· 5 defective screws. If a random sample

of 3 screws is
selected at random. What is the probability that the number of defe
ctive screws in the
s~mple 1s 2.
Solution:

Let X = No. of defective screws

then P( x = 2 ) ==
 Ch # 1 O Spec/a/ Probability Di.\trilmti,m \
-~~- ~ - - - - - - - 343
Poisson Probablllty Distribution
The formuln of Poisson Probability hstril.,ut1on is given by
C "'. ,,_,i
P(x) = for x = 0, ] , 2, ............n
xi
where x = No. of successes
m = average number of successes in a given time interval or space
e = 2.71828
Poisson distribution is often used to compute probabilities for random variables
~ uted oYer time and space. ·
Examples of Poisson random variables are the number of calls coming into a
telephone switchboard during a fixed peri9d of time, number of typing errors per page
made by a typist, the number of automobile accidents at a particular intersection in one
year~ the number of defects in a square yard carpet etc.
Example 1
A typist makes 1 errors per page on the average. What is the probability that she
makes (i) no error (ii) J errors (iii) 2 or more errors, on the next page she types.
Solution
Since m = 3 and x = No. of errors per page
e- m . m . e-3 • 3.t
X

then P(x) = == - --
x! x!

e-3 . 30 = e- 3 = 0.0498
(i) p (no error)== P( x = 0) = 0!

-3 33
(ii) P (3 errors)= P( x = 3) = e · = 0224
3!

(iii) P (2 or more~rrors) == P( x ~ 2) = 1 · [ P( x - 0) + P( x = 1) 1
3 1

1_ e 3 · 3° + e•- · 3 ] = 1 - 01992 i: 0.8

[ O' 1!
 Statistics Probkms & Practice . 1·
~..\.~~
-
,\ u ,'"wt.1in intersection results in 5 traffic accidents per month. Vvb.
:t\, \$\
,, , '--,,~ !'- ht\ that ex net ly 4 nccidcnts will occur at this intersection in any . at
' ~~

Since m = 5 and x = No. of accidents per month

-m X -s x
then P( x ) = e ·m e . 5
x! x!

P(x = 4)

Example3
= e -5 . 54
4!
- 01755
•
The average number of customers at a window of a certain bank per minute during
banking hours is two. Find ·the probability that at _any randomly selected 3 minute
interval, 4 cust(?mers appear. ·
-Solution:

x = No. of customers appear per 3 minute interval

m = average no. of custo1ners appear per 3 minute interval
= _6 (i.e. 2 customers in on~ minute= 6 customers in 3 minutes]
6 64
~nd P(x = 4) = e- • = 01338
4!
I

Poisso~ Approximation to the Binomial Distribution

Binomial distribution becomes as the Poisson distribution,
when;
(i) n (No. of trials) becomes very large i.e. n ;2: 20
(ii) p ( Probability of success) becomes very small i.e. _P ~ O.OS and

(iii) np =m
 £._~~f#~IO~S~p~ec~w~l~P~ro~bn~b~il~~~•D~l~Mr~/l~w~tl~ot~•s _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 345
Example 1

The probabilit y that a person dies from a certain respirator y infection is 0.005. Find
the probabilit y that exactly 2 of the next 1000 so infected will die.

Solution

p = 0.005 and n = 1000

therefore, m = np = (1000) (0.005) = 5

-5 5x 5
52
Hence P(x) = e · and P( x = 2 ) = e- • = 0.08

•·
x! 2!

Continuous Probability Distribution

So far we have been dealing with discrete probability distribution i.e. distribution
in which the random variable x takes on integer values such as 0, 1, 2 , 3, .......... But when
we deal with random variables like heights and weights, we find that such random
variables can take any value in the given interval a ~ x ~ b . Such random variables are
called·continuous random variables and their probability distributions are accordingly
known as continuous probability distributions. ·
The most important distribution defined on a continuous random variable is
Normal Probability Distribution.

Nonnal Probability Distribution

The equation of the normal probability distribution is

. ·1 _.!. (~)2
f(x) = ----~- e 2 er ' -ex> ~ X ~ ex>
.J2-i.cr .

where µ = Mean , cr = Standard Deviation

7t (pi) = 3.14159 and e = 2.71828
 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _~S~w~u~sd~a~P~r~ob~k~m~s~&~Pr~ac~~c~e
346
Graph (or shape) of the nonnnl distribution
is displayed by the curve; where
the total area under the curve is unity.

µ = Mean
keas under the Normal Curve
The curve of any contin uous probab ility distrib ution is constructed so that the area
under the curve bound ed by two ordina tesx = x 1 and x = x 2 equals the probability that
the random variab le X takes a value betwee n x = x 1 and x = x 2
Thus, for the normal curve in the given

figure, the Pr (x 1 < x < x 2 ) is represented by

•
the area of the shaded region.
i.e. Pr(x 1 < x < x 2 ) = Area of the shaded regio.;61.,....,._
To determine the area or probability of.an in
mean =µ and standard deviation = cr, _the foll
{i) First we conven x values into z values by
X - µ
z =
cr

where z is called standard normal vari

standard deviation = 1
Hence we have Pr(x 1 <
(ii) Determine the probability or area for
from the "Normal Area Table" encl
(iii) Subtract the area below the value.
Z = Z2

Therefore Pr (x 1 < x < x 2 ) = Pr.

Note: Normal area table enclosed at the
standard normal curve correspon ·
of Z = z orP(-oo < Z < z)orar
£hJ!. 10 Special Probability J)istl'ib1,ti011 s
· ----~-- 347
Example 1
Given a normal distribution ~ith P ~ 40 "nd o· = 6, find the probability ( or a 1ea) tllat
x assumes a value (i) Below 4 2 (u) Above 27 (iii) Bctwccl1 42 an<l 5 J
Solution

Here µ == 40 and cr = 6, then

(i) P( below 42) = P(x < 42)
Now transform the x-value into the value of z
x ·- µ x - 40
i.e. Z = _ _.;.... = - --f._'d for x = 42
•- cr 6 '

z= 42 - 40 = ~ = 033 µ =~O 42
6 6
then P(x < 42) = P(z < 0.33) = 0.6293 ( from Normal area tabie)
(ii) P( Above 27) = P(x > 27) = 1 - P(x < 27)

Since z = x - µ = x - 40 and for x = 27

(j 6

z= 27 - 40 = _ 2.17 27 µ = 40
6
therefore P(x > 27) = l - P(x < 27) = 1 - P(z < - 2.17)
= 1 - 0.0151 = 0.9850
(iii) P( between 42 and 51) = P(42 < x < 51) = P(x < 51) - P(x < 42)
x- µ X - 40
smcez = - - = - - - for X = 51,
cr 6
z =51 - 40 = 1.83
6
µn40 42 51
42 - 40
and for x = 42 , Z = = 033
6
therefore l>(x < 51) = P(z < 1.83) = 0.9664
P(x < 42) = P(z < 0.33) =0.6293
Hence P( Between 42 and 51) = 0.9664 - 0.6293 = 0.3371
 __cn_·_ce
_..,._..., _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _....;:S.~t~at'f!_ls~ticsr:_!...!h~o~b~/.e'!!.:."'~s=&:.,;:.P..;..ra

348 -
Example2
2 nd 18
Given a normal distribution withµ -12 and Ci= fiTh ea between 6 an d
1
(a) The area below J3.5 (b) The area above O (c) ear

Solution

(a) P(x < 13.5) == P ( z <

135 - 12
2
J
== P(z < 0.75J = 0.7734
(b) P( x > JO)== 1 - P(x < 10)
==1-P ( z<
IO - 12
2
J
= 1 - P(z < -LOO)
= 1 - 0.1587 = 0.8413

(c) P( 6 < x < 18) ==-= P(x < 18) - P(x < 6)

=P z <
18 2- 12) - p ( z '"'
6 - 12)
( 2

= P(Z < 3.00) - P(z < -3.00) 6 µ=12 18

= 0.Q987 - 0.0013 = 0.9974
Example3
The burning time of an experimental rocket is a random variable having the normal
distribution withµ - 4. 76 seconds and <1 = 0.04 seconds. What is the probability that
this kind of rocket wi 11 bum.
(a) less than 4.66 seconds (b) more than 4.80 seconds
(c) anywhere from 4. 70 to 4.82 seconds
Solution

Sinceµ = 4.76 and cr = 0.04, then

(a) P(x < 4.66) = p( z < 4.66 -
0.04
4.76) 4.66 µ =4.76

= P(z < -2.50) = 0.0062

- ----- ---- --
fl! !!J_~o:Sp:ec:ia-=I~Pr~ob:a:bl~llty 1Jl~,,t:rll~u,~tl,~m~,,~=====-~ ~~~~~-="
7
(b) P(x > 4.80) - l - r(x · 4.80) 349
_ I _ p ( z < 4.80 4.7<> )
0.04
= 1 - P(z .,- 1.00)

1 - 0.8413 0.1587
(c) P(4.70 < x < 4.82) = P(x < 4.82). P(x .!' 4.70)

=p ( z< 4.76)
4.82- -- - - -P ( z <
4.70--- -
4.76)
-
0.04 0D4

= P(z < 1.50) - P(z < 1.50) 4.70 µ • 4.76 4J2

= 0.9332 - 0.0668 = 0.8664

Examp1e4
Given a normal distribution with µ = 100 and a • 5, ftnd
(a) The area below 107 (b) The area above 89.5

(c) The area between 94 and 103

(d) The point that has 80% of the area below it
(e) The two points containing the middle 75% of the area

Solution

(a) P(x < 107) = P ( z <

107 - 100)·
5
= P(z < 1.40) = 0.9192

(b) P(x > 89.Sj = 1 - P(x < 89.5)

=1-P ( z <
_s95_- _1oe)
s
= 1 - P(z < -2.10)
= 1-0.0179 = 0.9821
350 ----====~ ~===--c
-=-- -=• • - - - - - =Sta!J!tlcs Problems & Pral'fl-e
~
(c) P(Q 4 x J 11) P(x I O:l) P(x 94 )
0.6106 ,
Wl 100) 94 S I 00 J
5
P( (1 60) P(1 I 20) 94 µ = 100 103

0 I l 5I O 6100
( u) et •k~ be the point that has 80% of the area below it as shown in the
gJ\ en fig ure.
In other\\ ords P(x < k) = 0.80

then P ( =, k - 100 l) = 0.80

5
k - 100 µ = 100 K
an d - - - = 0.84(fromnormal table)
5

k = 104.2

(e) Let k1 and k 2 be the two points containing the middle 75% of the
't{ area as shown in the given figure
then P (x < k 1 ) = 0.125

p( Z < k I - 100) = 0.125

5
k1 - 100
and = - 1.15 ( from normal table)
5
and k 1 = 94.25
Al o P (x > k sub size 7 2 ) = 0.125
P(x < k2 ) = 1 - 0.125 = 0.875

"' P ( z < k 2 ~ JOO ) - 0.87 S

and ~ : 100
5 115 ( from normal table)

k, = 105.75
 CJ, # 10 Special Probability Distributio11s
----__,;,._;;....;..;.;:..::::..:.:. ::.z..!::~~~------------351
Normal Approximation to Binomial Distribution
When n is large and p is not very close to O or to 1, then binomial distribution
approximates to normal distribution having µ = np and cr = .[npq where z = x - np
✓ npq
is the standard normal variable
Note: It is to be noted that when we use Normal distribution (i.e. continuous
distribuuon) to approximate a discrete random variable. So we make a continuity
correction by adding or subtracting half the unit of the measurement to each value
of the discrete variable.
For example:
Ifx is bionomial (i.e a discrete) variable, then x can take on any of the values
0, 1, 2, 3, 4, Therefore, the values of discrete variables are transformed as
P(x = 4) transformed to P(3.5 < x < 4.5)
P(x < 4) transformed to P(x < 3.5)

P(x ~ 4) transformed to P(x < 4.5)

P(x ~ 4) transformed to P(x > 3.5)

2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8,5 9.0 9.5
P(x > 4) transformed to P(x > 4.5) µ
P(5 s; x < 8) transformed to P(4.5 < x < 8.5)
P(S < x < 8) transformed to P(S.5 < x < 8.5)
P(5 s; x < 8) transformed to P(4.5 < x < 7.5)
P(5 < x < 8) transformed to P(S.5 < x < 7.5)
As shown in the following probability Histogram.
 0
- - - - - - - = - -"'•=-- - - - - - - - - - - '~_ta__tl_stl_cs_~_r,~ob;.:.:;le;.;,:,;m:::..s.:::,&~Pra
~
,J

352 -
Example ~
A Com s tos~cd 400 times. Use the nonnal curve cipproximation to find th e
1
probab111t, of Jbtaming
(a) Between 18S and 210 heads inclusive (b) Less than 209 heads
(c) At least 200 heads (d) Exactly 205 heads
Solution
1 1
n = 400 and p =- Then µ = np = ( 400) - = 200
2 ,------ 2
and cr = .J npq = /400 x ! x ! = IO
v 2 2

where z = x - np = x - 200
.Jnpq 10

since the data ( No. heads) are discrete and measured in . t

continuity correction. m egers we need the

P(l85 ~ x ~ 210) = P(1845 < X < 2105)

=P(x < 210.5) - P(x < 184.5)
2105
= P( z < - 200 ) _ ( 1845 - 200 )
10 p z <
10
= P(z < 1.05) - P(z < -1.55)
= 0.8531 - 0.0606 = 0. 7925
(b) P(x < 209) - P(x < 208.5)

=p ( z < 2085 I~ 200 )

= P(z < 0.85) = 0.8023

(c) P(x ~ 200) = i P(
- X < 200)

= l - P(x < 199.5)

 Ch # 1 O Special Probnblllt•' Dh,trlh11tfon \;

~--------353
I•p( Z < 199'\ O ;tVO )

1 - P(z < -0.05)

1 .. 0.4801 0.5199
d) P( \ = 205) = J->(2045 < x < 2055)

= P(x < 205.5) - P(x < 204.5)

_p(
- 2055
Z< - -- -200)
10
- - p( Z< 2045 - 200)
10
= P(z < 0.55) - P(z <0.45)
= 0. 7088 - 0.6736 = 0.0352
Example2

A hockey player hits on 60% successful shots in penalty comers. What is the
probability that he makes less than 50 successful shots, out of 100 opportunities to be
given to· him in various matches to be played next month.
Solution

n = I 00, p = 0. 6, q = 0.4
Then µ = np= (100)(0.6) = 60
andcr = ~ = ✓100 x 0.6 x 0.4=4.9

P(x < 50) = P(x < 49.5) =P ( z < 495 - 60)

49

= P(z < -2.14) = .0.0162