15.

RANDOM VARIABLE AND PROBABILITY DISTRIBUTION

If an experiment is conducted under identical conditions, values so obtained may not be similar
Observations are always taken about a factor or character under study, which can take different values.
This factor or character is termed as variable. The observations may be the number of certain objects
or items or their measurements. These observaticns very even though the experiement is conducted
under identical conditions. Hence, we have a set of outcomes of a random experiment. A rule that
assigns a real number to each outcome is called random variable. The rule is nothing but a function of
the variable, say, X that assigns a unique value to each outcome of the random experiment. It is clear
that there is value for each outcome, which it takes with certain probability. Thus when a variable x
a
takes the value x, with probability p,(i=1,2,
variable or a variate.
3,.. n), then Xis called random variable or stochastic

1.6. DISCRETE RANDAM VARIABLE

A random variable X, which take
can only a finite number of values in an interval of the domain is
called discrete random variable.
 BASIC PROBABILITY
31
Example:
1. Number appearing on top of a die when it is thrown.
2. The number of telephone calls received per day.
3. Number of mistakes in a page.
4. Number of defective items in a lot.

1.7. DISCRETE PROBABILITY DISTRIBUTION

Ifa random variable x can assume a discrete set of
values say x, x2, x , with respect to probabilities
D,PpP such thatpi *P2*.PIie., 2,pt =

1then occurrences of values x, with respective

i=1

probabilities p, is called the discrete probability distribution of X.

For example, In a throw of a pair of dice the sum
(X) is discrete random variable which is an integer
between 2 and 12 with probabilities P(X) given as

2 3 7 8 9 10 11 12
2 6 4
PCX 36 36 36 36 36 36 36
3
36
1
36
This constitute a discrete probability distribution.
1.8. PROBABILITY FUNCTION OR PROBABILITY MASS FUNCTION
(pmf)
Probability function of probability mass function (p.m.f.) of a random variable Xis mathematical
function p(x) which gives the probabilities corresponding to different
possible discrete set ofvalues say
X1 2 *3 . *, of variable x.

i.e., p)= p(=x)

probability that and on variablex assumes value x;
The function p (x) satisfies the condition.
Pa)20
() Ep)=1

11.9. CUMULATIVE DISTRIBUTION FUNCTION (DISTRIBUTION FUNCTION)

I fX is a random variable then P(Xs x) is called the cumulative distribution function (cdf) or
distribution function and is denoted by F(x)
Fx)= pX<x)

1.10. EXPECTATION OF A DISCRETE RANDOM VARIABLE

Ifx is discrete random variable which assumes the discrete set of valuesx,.Xp,with respective
probabilities p, P2...p, then the expectation or expected value of x is denoted by E(X) and defined as
ECX) = x P t x 2 P2 t X3 P3 «. **,Pn
 32. PROBABILITY AND STATIS

Similarly the expected value of X2 is defined as E(x2)= 2 P

Properties:
1. IfXis a random variable and 'a is constant then
() E(a) =a
i) E(a) =aE()
(i) E (X-u) = 0
2. Ifr and y are two random variables then E(X+ Y) = E(X) + E(Y).

3. E(X )=E() E(Y ifXand Yare two independent random variable.

4. ify=ax +b where a and b are constant then E(Y) = aE(X) + b.

1.11. VARIANCE AND STANDARD DEVIATION OF DISCRETE RANDOM VARIABLE

The variable of discrete random variable Xis
expected value of(X-u) where u is mean of varaible
Var X V(x) =E (X-u)2
2-u*
i=l

= p (x-p)2

=
2pr2 + Zpu2-22pux
pr+p-2px
Epr3+ 2-2
pr-2
= EX)- [ E P

The standard deviation (SD) of a random variable x is denoted by s, then

SD ()= v(x) JE(xy -(E(X)*
=

sOLVED EXAMPLES
Example 1.41. () A pair of two coins is tossed, what is the
expected value?
(i) A pair of dice is thrown together, find the
expected value.
Solution. (i) Expected value or mean value
E(X) =
H

PA (Here X is a discrete random variable)

In tossing of two coins, probability distribution is represented in tabular form as follows
BASIC PROBABILITY. 33
X 2

P(x) 4

E)x0x1*x2=1
1 11
As the probability of getting no head, one head and two heads is respectively

( In a throw of pair of dice the sum (X) is a discrete random variable which is an integer between
2 and 12 with the probabilities as given below:

x2 34 6 8 10 11 12
3 4 5
336
PCX) 36 36 36 36 36 36 36

Expected value - E ( ) = 4 2 t 3 4 +1 0 l l 1 2

252
-36 7
Note. The variance in each of the above cases is given by
- PA-w = Epx' -u'
In the tossing of two coins, we have

p.10o*.
Variance = = p - f = - 1 - 5
In case of a pair of dice, we have

p 4916253649
64+81+100+
36 36 36
121+144
36

=4+18+48+100+180+294+320+324+300+242+144)
(1974)=329
36 6

Variance g--(7=
Standarddeviation= g 35
 PROBABILITY AND STATIS
34

1.42. A random variable X has the following distribution

Example
2 3
X -2 -1 0
2k 0.3 k
0.1 0.2
P(X)
Determine: () k, (in Mean, (ii) Variance.
Solution. () Since EPr) = 1 |

0.1+k+0.22k+0.3+k = 1
0.6+4k= 1
k= 0.1
random variable Xis
Now the probability distribution ofthe
2 3
X -2 -1 0
0.2 0.3 0.1
0.1 0.1 0.2
PX)
(i Mean 2xp(r)
= (-2) * (0.1) +(-1)* (0.1) + (0) x (0.2) + (1) x (0.2) + (2)
x
(0.3) + (3) * (0.1)
= 0.8

(iii) Variance Expx)-Exp(x)]?

= (-2)X0.1)+-)(0.1)+(0)(0.2) + (1) (0.2) +(2) (0.3))
ao + (3) (0.1)-(0.8)2
2.16
contains 8 items of which 2 are defective. A man selects 3 items at
Example 143. A bag
random. Find the expected number of defective items he has drawn.
Solution. The expected mumber of defective itemscanbe zero defective, one defective, two defective
items. Thus, a random variable may take values 0, 1 and 2.

C6,3)x C2.0)
Now P PCX=0)= C(8,3)
623!x5! F
20
3!x2! 2!8! 56

P2 PX=1) = C(6,2)xC(2,1)
CC8,3)
6! 2!3!x 5! 30
4!x1!IxI 8! 56

P3P(X=2) =C(6,1)xC(2,2) 6! 2! 3x5 6

C(8,3) 5 2 8! 56
Hence, the expected number of defective items drawn is
E)-P tP2* tp,*
20014x2 -
56 S6 56 56
BASIC PROBABILITY
35
Example 1.44. A player tossed two coins. If two heads show he
wins 4. If one head shows he
wins 2, but if two tails show he pays 7 3 as
him.
penalty. Calculate the expected value of the
game to
Solution. Here Xtakes the values 0, 1,2

Also P P(X= zero head) = x

P2 PR= one head) =

P3 P(X= two heads) =x

Also when X =
0
x=R 3
X =1 x=72
X =
2 x^= 4
We want to find out

Ex)= PX1tP%*P3x 3)* = 1.25

Hence, EX)= T 1.25.

Example 1.45. Find the mean and variance of

uniform probability distrubutionf(r) =
for
x=1,2.n
Solution. X1
n

P(X) n

Mean E(X)=++ n

1+2+3.n
n

1 nn +)n+1
l 2 J2
Variance EX)-[E(X)P
F+2 t n
+
n(n+1)(2n+1)
6n

n+ 2n+l_(n+1)
23 2
36. PROBABILITY AND STATISTIC-

Example 1.46. For discrete probability distribution.

7
0
k
2
2k
3 2k
4
3k
5
2kK
6
7k2 +k

Determine: () k, (ii) mean, (iin) variance, (iv) smallest value ofx such that P(X Sx)> ;.

2fx) = 1
Solution. ()
0+k+ 2k +2k+3k+ k2+2k2 +7k2 + k=1
10k2+9k-1 = 0

1
10k -
1 or

k = 1 since probability can never be negative.

4 5 6 7
0 1 2 3

1 2 3 2 77
0 10 100 100 100
10 10

(i) Mean 2xf)

0+1x2x+3x+4x+5x+6x2
100
+ 7x 100
= 3.66

(iii) Variance = E(X) - [E(X)1

-0+Px+2 x+3 x+4x+5x+6

10
x+
100
7x -(3.66
10 10
= 37.7

Px 0)f(0) =0
(iv)
P(rS1) = f0) +f(1) =0.1
P(xs2) = 0+0.1 +0.2 0.3
P(xs3)= 0.3+0.2 0.5
P (x4) =0.5 +0.3 =0.8
Smallest value of X such that P(KSx) 20.5 is 4.

Example 1.47. Two dice are thrown. Let X assign to each point (a, b) in S the maximum of its
number ie, X(4, b) max (a, b). Find the probability distribution of random variable x with
X(S) = {1, 2,3, 4, 5, 6 .

Solution. When two dice are thrown sample space Sis

(1,1), (1,2), (1,3), (1,4), (1,5), (1, 6)

(2,1), (2,2),(2,3), (2,4), (2,5), (2, 6)
SS-
(3,1),(3,2). (3,3),(3,4).3.5),(3,6)
(4,1).4,2),(4,3),(4,4),(4,5),(4,6)
(5,1).(5,2), (5,3),(5,4),.5,5),(5,6)
(6,1), (6,2), (6,3), (6,4), (6,5), (6,6)
BASIC PROBABILITY 37
From the data X[1, 1] = max (1, 1) = 1

PX=1)= 1/36
X[(2, 1) (2, 2), (1, 2)] - 2, P(X=2) = 3/36
X[(1, 3) (3, 1)(2, 3), (3, 2) (3, 3)] = 3, P(X=3) =5/36
X[(1,4) (4, 1)(2,4), (4, 2) (3, 4) (4, 3) (4, 4.)] =4, P(x = 4)] = 7/36
X[(1, 5)(5, 1)(2, 5) (5, 2) (3, 5) (5, 3), (4, 5) (5, 4), (5, 5)] = 5, P(X=5) =9/36
X[1,6) (6, 2) (2, 6) (3, 6), (6, 3) (4, 6), (6, 4), (15, 6), (6, 5), (6, 6)] = 6, P(x = 6) = 11/36
Probability distribution to random variable Xis
X 2 3 4 6

P(X) 9 11
36 36 36 36 36

Example 1.48. A fair coin is tossed until head or five tails occurs. Find expected number of
tosses of the coins.
Solution. Probability of getting head=1/2-p
Porbability of getting tail = 1/2=q

X 1 2 3 4 5 6
Outcome H TH TTH TTTH TTTTH TTTTT
Probability
The expected number of tosses are
Ex)= 2xP;
1p + 2pg+ 3.gp + 44p +
=
5q'p + 6q

-1)))
= 1.9687 =2

Therefore expected number of tosses are 2.

Example 1.49. Two cards are drawn successively with replacement from a well shuffled pack
of 52 cards. Find the mean and variance of the number of kings.
Solution. Let Xbe the random variable. The
X= Number of kings obtained in two draws
Clearly, Xcan assume the value 0, 1,2
4
P (drawing a king)= 52 13
P (not drawing a king) |--12
=13 13
P(X= 0)= P(not a king in the Ist draw and not a king in 2nd draw)
144
169
 PROBABILITY ANDNO STAn
STATISTICs
38.
in the lst drawn and not a king in the 2nd draw)
PX=1)= P(a king lst draw anda king the 2nd d-
in
orP (not a king in the drawn
24
- 169
in the 2nd draw)
P(X=2)=P (a king in the 1st draw and king
a

Hence, the probability distribution is given by

2
X=x; 0
144 24
Pi 169 169
169

Mean, (140
169
Variance, o p,x-

169

Example 1.50. An industrial salesman wants to number of units the sells

know the average
per sales call. He checks his past sales records and comes up with the following probabilities?

Sales in units 0 1 2 3 4 5

Probability 0.15 0.20 0.10 0.05 0.30 0.20

What is the average number of units he sells per sales call1?

soluion. Ihe sales wants to known the average number of units be sells per sales call. Thisisthe
same thing as saying that the wants to know the expected value of each sales call, where a sales callis
the random variable The
X. expected value is calculated the formula by
ECX)= PXtp2 *2 t p3 *3t
=0.15 x 0+0.20 1+0.10 x x 2+0.05 x 3+0.30 x 4+0.20x5
=
0+0.2+0.2 +0.15+ 1.0=2.75
Thus he would expect to sell 2.75 units on
each sales call.
Example 1.51. A die is tossed twice. A 'success' is
Find the variance of the number of successes. getting an odd number on a random to
ss.

3
Solution.
P(success) 62
and P(failure) =1-5-
BASIC PROBABILITY 39

PX=0)5*
P(X=1) = -
PIX-2)
Hence, the probability distribution is given by
-
X

1 1
P()
4

,a-Exo+ 4

Variance, o-p,x2-H=

Example 1.52. Find the expected value of the absolute difference of upturned faces in the
experiment of tossing of two dies.
Solution. Abs. DIf. Total no. of cases

Pr)
0 (1,1). (2, 2), (3, 3), (4, 4), (5, 5), (6, 6) 6/36
(1,2) (2, 3), (3, 4), (4, 5) (5, 6)
(2,1), (3, 2) (4, 3), (5, 4), (6, 5) 10/36
(1,3).(2, 4, 3,5), (4, 6), (3, 1), (4, 2) (5, 3), (6, 4) 8/36
3 (1,4), 2, 5), (3,6), (4, 1), (5, 2) (6, 3) 6/36
4 (1, 5), (2, 6), (5, 1) (6, 2) 4/36
5 (1,6) (6, 1) 2/36

Hence, Ex)= 2p(x)

X=0

2
=

70 35
36 18
Cxample 1.53. If X and Y are discrete random variables and K is a constant then prove that:
() ECX+ K) =
E (X) +K and (i) E (X+ ) =
E (X) +E (Y).
Solution. Given Xand Yare discrete random variable and K is a
constant.
Since EC) = P a ) and P =1
i=l
 PROBABILuTY AND ATISTICS
40

ECX+K) (X+K)»,
Now

P +KP =
E(X) +K-1
i=1

= E(X) + K

E(X+K) E(X) +K

X+- 2a*y)p, 2AP*»P

i=l
-

i=l i=l

= E(X) +E(Y

E(X+ Y) = EX) + E(n

EXERCISE 1.3
11.40. CHEBYSHEV'S INEQUALITY
Let X be a random variable with E[X] = u and var [X]= o?.
Then for any constant c > 0,

P{| X-p |2c} o/c or P {|X-p|<c} 21-(G/). ..(i)

Proof. By Markov's inequality, we have

P[X-u 2 s EICX 2--o

Hence, P[X-|2c]s olc2. [:: 22a |x|2a for a > 0]
To prove the second part, we notice that

P[I X-|<c]+ P[|X-u|2 c]=1

P[1 X-I<e]=1-P[X-u|2c]21-(62).
oIeis called Chebyshev's inequality bound for P [|X-m |2c
Remark 1. Chebyshev inequality can be restated as

P[X-ELX]|<e]2 1-var
c
[X].
Remark. 2. IfX is a random variable with finite variance, then

PIX-H 2r o]2orP [|X-u|<r o] 2 1- for eachr>0.

r*
The above results follow by taking c =r o in (i).
 PROBABILITY AND S
Remark 3. From the above
inequality, we have TATISTICE
P[u-ro <X<p+ro]2 1-5
It follows that the
probability that Xfalls withinra units of u is greater than or equal to 1-
Remark4. The Chebyshev s inequality
of a random variable and its
gives a bound for the probabilityof an event given in ter.
mean and variance. Further this bound does not depend on the
ofX. distribui.

sOLVED EXAMPLES
Example 1.92.IfXis a random variablesuch that E[X]= 3 and EX"=13, use the
Chebyshev',
inequality to determine the lower bound for P-2 <X< 8].
Or
If X is a variate such that EX]
3, EX"] = =
13, show that
P-2<X<8] 2 21/25.
Solution. We know var [X] =
E[X]- {E[X]}
=
13 -9 =4 0
=2.
Now P-2<X<8] = P[-2-3 <X-3 <8-3]
=
P-5<X-3 <5]
P[IX-3 |<5]21 -(2/52). by Chebyshev's inequality
Hence, P-2 <X<8] 2 21/25 i.e., 21/25 is a lower bound for P[-2 <X< 8]
Example 1.93. Variate Xtakes the values-1,1,3,
5 with associated probability zT7:.
Computep=P[|X-3|21] directly and find on upper bound top by using
Chebyshev's inequality.
Solution. ELX= - = 3,

ELX = 1. 43

o= var X- E[X]- {ELX}2= 9

By Chebyshev's inequality,
=
we have

p=Pllx-3|s 2 ( c=1)
Hence, an upper bound to p is 16/3.
Actual
p 1 -P[lx-3|< 1]=1 -P- 1 <X-3 <
1]
-
1- P[2 <X< 4] 1
= -

P[X= 3] =

1- -
BASIC PROBABILITY 87
Example 1.94. If Xis the number scored in a throw of a fair die, show that the Chebyshev's
inequality gives Pl| X-4|> 2.5] <0.47, while the actual probability is zero.

Solution. We have j =E[X]= (1+2+3 +4+5 +6) = -3.5,

91
ELX=12+22+3+4?+S3+6)-
o- var[X]-ELX]- {ELX]}2- 12
By Chebyshev's inequality, we have
35
P[X-HI >2.5] 5047.
12 (2.5)
Actual probability =

P[|X-4|>2.5] 1-P[| X-3.5 |S2.5]

=

=
1-P-2.5 +3.5 X<2.5 + 3.5]
=

1-P[1SXS6] = 1-1 =0.

Example 1.95. A discrete variate X can assume only the values x=1,2,3,...with probabilities
1
2. Show that Chebyshev's inequality gives PlIX-21s2]>,while the actual probability is
15/16.

Solution.

(1+24+342 + 443.), (A = ;

(1-42=2.

E'1
-+44+94.. d+A)0-A)" =6.
=

o var [X] =E;X]- {E[X]}2 =6-4 2.

By Chebyshev's inequality, we have

P[IX-2 2]> 1-=1-5 ( c 2, o2 =2)

PEIX-21s2]>
The actual probability is given by
P[X-2|s2] P[-2 s X-2s2] =PO SXs4]
 88
PeOBABILTI AMD GTAte

P1]+ P12 + P13] +P14], PMx) 2

15

Example 1.96. Let X have the pd.f.:

- 3 <x<3
fAx)23 elsewhere.
0,

3 and compare it
al
with the upper bound obtained
Find the actual probability PI| X-p|S

by Chebychev's inequality.
Solution. We have

ELX- J d =
0,

EL- d -

63 -3/5) - 1.

and u =0.
a= var [X]=E[X]- {E[X];2=1
have
By Chebyshev's inequality,
we

=1.c=
The actual probability is given by

( 0 , G =1)
P -u/2

-PxI
dx

23
23,3-132-V3
/2'222
Hence, the actual probability is (2 3 2 , whereas the upper bound obtained by Chebyshev
inequality is 4/9.
BASIC PROBABILITYy 89
Example 1.97. The rainfall Xin a certain locality is a normally distributed random variable
with mean 40 cm and variance 4 cm2. Find a simple upper bound on the probability that rainfall
in a particular year will exceed the mean by 5 cm.
Solution. We have u = 40 and o2 = 4.

By Chebyshev's inequality, we have

P[IX-42c]s o/c2 or P[I X-40 |2 5] 4/25.
P[I X-40 |25] 0.16.
The exact value of this probability is
P[X-40|25]= 2 P[X< 35]
X -40 35-40
=2
2 -
where Z= (X-40) - N(0, 1)
=
2 y(-2.5) =
2 [1 -y (2.5)]
2 [ 1 - 9938] =2 x .0062 = .0124.

Thus an upper bound to the probability is 0.16, which is totally different from the exact probability
.0124.