ntroductionto Statistics

177
Variance
fXfollows Poisson dictribution with parameter 2then
e poitudhuib
EX)= x!
x=0

=x
X=1 x(x - 1)!

=e X=1 (x- 1)!

siorns
=e a1++ 2! t...

ariance
VIX) = E(X) - {E(X)}2 . (1)
Tnoniz
Now, E(X*) = E(X(X - 1) + X},v
= E{X(X - 1)}+ E{X}2u Helz SW 9Ttsid(2)
t9mun t sionst
ó x(x- 1) "2
Again E(X(X - 1)}= X=0 x!

=e X=2
x(x - 1) x(x - 1) (x -2)!

=e 'x=2 (X -2)!

=e2 1 +2+ +..

2!

e
= 'e ..3)
zwollot XT
fudiitaib lsiianie
From (2) and (3) we get
E(X*) =22+ E(X)= 2
i. from (1) we get
V(X) = 22+-2
distribution mean = variance,
MhichRemark :(1)
is the important
For the Poission of this
characteristic
distribution.
 Introduction to Statistics 179
178

Standard deviation = /
Standard Distributions An
Solution:
(2 Hare since n = 100 is very large and p -501 is Very small
Some exanples
of the Poisson distribution thereforeto find the approximate probability we shall use here
() Number of the deaths by horse kicks or Poisson probability Ilaw
(ii) Number of printing mistakes in a
book.
particularSnakepagebiteofs a Hete 1= np= 100x=1
50
(i) Number of faulty blades in a packet of 100 e2
(iv) Number of suicides reported in city etc. .:. P(X =x) = ;x=0, 1, 2....100
x!
Example -11 P(1<X S 3) =P(X =1) +P(X =2) +P(X =3)
From a population consisting of defective and e'2 e22
items, 100 items are drawn at random. Find the non-defectiveof 1!
+
2! 3!
getting 2 defective items. Given that the population probabi lty
consists of =e()=0.388
1% defective item.

Solution: Example - 13
Here since n = 100, is very. large and p= 1% = 0.01 is yery If xfollows Poisson distribution with E(X*) = 6, find E(X)
small therefore we shall use here Poisson probability law. Let Solution :
Let E(X)2
= O i sonu `nt ai ei
Xdenote the number of the defective items. roudsb noala9
Now V(X) = E(X*) - {E(X)P
1
Here 1= np=100 x100 =1 : here E(X) = V(X) =2
’ N+ -6 = 0
.. P(X= x) = ;X=0, 1, 2... or, 2 2
x! 2 = -3
cannot be negative .:. 2=2
i. P(X =2)= e (0.27)} e = 2,7 Moment Generating Function
2! 2
= 0.0196
X=0 x!
Example 12
If Xfollows Binomial distribution with probability mass (e'2y
function X=0

1 )100
14+ .e'2,(e2)+ +
bist L 11
X=0, 1, 2,..100
find the alqmsd
approximate value of P(1sXs3) =e(1-e')ayEAA
POISSON DISTRIBUTION 113

Example 1. X follows the Posson distributon and P(X) = 1) =P(X = 2) ; find

(AHSEC2000)
PX = 4).
Solution. If X ~ p)

PX = x) x!
;x0, 1, 2, ....
Given, P(X = 1) P(X = 2)

. 1! 2!

2
=2

P(X = 4) =
4!
16
X.135
24
= 0.09.
= a. Show that
Example 2. Xfollows Poisson distribution and P(X= 0) = P(X = 1)
1 (AHSEC 1996)

Solution. Given PX = 0) = P(X = 1)

0! 1!
2=1
Also, P(X = 0) = a
= 1
0!

Example 3. If X has a Poisson distribution and P(X = 0) what is E(X) ?

(AHSEC 1997)

Solution., Given P(X = 0)

 AFIRST COURSE IN STATISTICS-I
114

e
2

-d =loge )
- l= log, 1 - log, 2
- 1= 0 - .693
-1 =- .693
E(X) = 1 = 0.693
= 2). Find
Example 4. A random variable X is such that P(X = 1) = 2. P(X
() PX = 0), (ii) P(X > 0).
Solution. Given, P(X = 1) = 2 P(X = 2)
-.2
0! 2!

=1

() P(X = 0) = 0!
= e-l =.368
=

() PÌX > 0) = 1
POISSON DISTRIBUTION TesE 115

Example 6. Write down the standard deviation of the following distributon.

e 4*
P(x) =. ;X=0, 1, 2, ... (AHSEC 2002)
x!
Solution. Here X ~ p(4)
S.D. = Va
oeeotcat stidie = v4 =2.8 oitihacsy
Example 7. The number of mistakes counted in 100 typed pages, of a typist
revealed that he made 2.8 mistakes on an average per page. Find the probability that
in a page typed by him.
(i) There is no mistake
(i) There are two or more mistakes. [Given, e2.8 = 0.061] [AHSEC 2001]
Solution. Given =2.8
Using Poisson distribution, the probability of x mistakes
,-2.8
P(X = x) =
e (2.8)* ;X=0, 1,2,...
a!
()) The probability of no mistake
2.8 (2.8)0
P(X = 0) e =

0!
0e28
= 061
(i) The probability of two or more mistakes
PX > 2) = 1- p(0) - p(1)
e2.8
=1-.061 - (2.8)teieob bsshneie
1
= 1-.061 - 1708
= 7682.
Normal Distribution:
Definition:
Acontinuous r. v X is said to follow
normal distribution is
s p.d.f is given by
182
Standard Distrbuton,
1 -c<X<c
f(x) = e
Odataib roa
- c< u<oc
O>0 0exbExD
ulcalled mean) and o (called s-d) are the parameters t
this distribution.
Remark:
If X follows normal distribution with mean uand s-d a
then it is usually denoted by X-N (u,o)
Statement of Derivation :
Normal distribution can be derived as a limiting form of the
binomial distribution under the following conditions. 90
(i) n, the number of independent
large ie n’o
Bernoulli trials is very
(iü) Neither p, the
q = 1 - p) is very small. probability of success in each trial nor

Standardised or Standard Normal Variate -

If X-N( , o) then Z= u-
is called asstandard normal
variate.The p.d.f of zis
z?
1
NZ)= e 2; -c<Z <«
7.5 PROPERTIES OF NORMAL DISTRIBUTION
1. The normal curve is bell-shaped and symmetrical about the line x = u and the
two tails of the curve on both sides of x = uextends to infinity.
2. The distribution is unimodel.
3. Mean = Median = Mode = u
4.
Normal distribution is a continuous probability distribution having two
parameters u and o².
5. The maximum ordinate lies at the mean ie., x = u. ouloe
6. The two quartiles Q1 and Q% are equidistant from the mean u and
Q1 = u - 0.6745 ¡
Q2= u + 0.6745 ß
7. The points of inflexion (the points where the change in curvature occurs) of
the curve are at x = ut o and x =u- 0.
8. O.D: M.D: S.D:: 10 : 12: 15
4
9. MD o and QD. o,
5
10. B =0 and Ba - 3
11. AIl odd order central moments are zero.
122
AFIRST COURSE IN STA
12. P(u-o< X< u + o) = 0.6826
P (u - 2o < X< u+ 2o) = 0.9544
P(u - 3g <X< u -3o) = 0.9973
f(x)

u-3o u-2 o u-o X=u u+0 u+2¡ u+3o

68.26%
95.44%
99.73%
Bample - 15
If X~N y o) find P( - Þ<X< u+o)
Solution :
P(u -o < X<p+o)
=P

[Since first we have to transform the given variate Xto

standard normal variate Z, therefore we have subtracted and
divided it by o]
X-!
=P(- 1<z< 1);

1
186
Standard Distribufis
(:: the area from -1 to 1 can be obtained
the area from (c- to -1) from the area (-0 to 1)]by
subtracting

6916

2=0 +I

N Z
(z)dz =1-[ (z)dz Here z, =1
=2 2)dz - 1 tib
=2 x0.8413 -1 [from table)
=0.6826
 Example 2. IfX isnòrmally distributed with mean 3 and
P(4 < X<6). standard deviation 2 find
Solution. Given, u=3
Þ= 2

P (4< X< 6) 4-3

-P<z<
DDIO
22
P(0.5 < Z< 1.5) /o
=(15) - o(0.5)
r Sb S s z=0.5 1.5
=
0.9332 - 0.6915
0.2417.
Example 3. If X ~ N (15, 16), find the probabilty that X is larger than 18.
Solution. Given u=15
g² = 16
o= 4 o e batvolse ylnobr
18 15)
P(X > 18) = 4
=
P(Z > 0.75) Z=.75

= 1-( < 0.75)

=1- (0.75) nhst: i
-1 - 0.7734
=0.2266isEY diors scsod iuio
 Example 5. The marks obtained in a Statistics examination are
anormal distribution with mean 80 and assumed to have
variance 25. Find the probability of a randomly
selected student obtaining marks betwen 75 and 85.
(AHSEC 2000)
Solution. Let X denote the marks obtained in the examination
u= 80
2 = 25
O= 5
X- 80
Z=
5
Probability that aa randomly selected student obtains marks between 75 and 85 is

P (75 <X<85) (75-

-P(2-80<z< 85-80
5 5
= (-1< Z<1)
=
0.6826.