5.16.

RANDOM VARIABLE

If the numerical values assumed by a variable are the result of some chance factors, so
that a particular value cannot be exactly predicted in advance, the variable is then called a

random variable. A random variable is also called 'chance variable' or 'stochastic variable.
Random variables are denoted by capital letters, usually, from the last part of the
alphabet, for instance, X, Y, Z etc.
Continuous and Discrete Random Variables
value within an interval,
A continuous random variable is one which can assume any
(i) the weights (in kg) of a group of individu
i.e., all values ofa continuous scale. For example
individuals.
als, (ii) the heights of a group of
variable is one which can assume only
isolated values. For example,
A discrete random
canno
of a coin is a discrete
random variable as it
of heads in 4 tosses
(i) the number
assume values other
than 0, 1, 2, 3, 4.
from a well shuffled
deck is a random
aces in a draw of 2 cards
(it) the number of
variable as it c a n take
the values 0, 1, 2 only.
 5.17. (a) DISCRETE PROBABILITY DISTRIBUTION
Let a random variable X assume values x,. a, xa . . , , with probabilities Pi Pr Pa
1.
2P
=
P
...P, respectively, where PX
=
x,) =p, 20 for each x, and p,+P2 +Pa*
Then
X
PX Pa P3 P
Pn
the discrete total probability of
is called probability distribution for X and it spells out how a

l is distributed over several values of the random variable.

s.17. (b) CONTINUOUs PROBABILITY DISTRIBUTION
Let X be a continuous random variable taking values in the interval ( - , ),. Let fr) be
a function satisfying the following properties:
( ) fx) is integrable on ( - , )
(u)fx) 20 for allx in (- o, o)

(iif)dk = 1

Then fx) is called the probability distribution (or density) function (p.d.f.) of the
continuous random variable X.
The probability for a continuous random variable X to fall in the interval la, b]is

Pla SX<b) =
fa) d
which is nothing but the area between the continuous curvey =fx), x = a, x =b and r-axis. The
curve y =
fr) is called the probability curve.

Taking b =
a, PX = a) =fx) dr = 0. Also,

P(a sXsb) P (a <X< b) P (a <Xsb) =P(a <X<b)

=
=

If X is a continuous random variable with p.d.f. Ax), then

Flx) = P (X < x) =
| flx) dx
distribution function of the con
is called cumulative distribution function or simply
function has the following properties:
tinuous rándom variable X. The distribution
(i) 0s Px) s 1, -

(ii) Fx) = flx) 2 0

P(Xsb)- PXsa) =Fib)- Fa)

a ) dx [flx) d -flds
=
=
(ii) Pla sXsb) =
5.18. MEAN AND VARIANCE OF RANDOM VARIABLES
a) Let g
PX P Pa Pa Pr
be a discrete probability distribution.

We denote the mean by u and define u =

p ( p 1
Other names for the mean are average expected value E(X)
or

We denote the variance by o and define o p,x, u

= -

If is not a whole number, then o p x -u

Standard deviation d = +
Variance.
The rh moment about the mean is denoted by 4, and defined as

4 , -u flx,) where fx,) =

p
Putting, r = 0, 1, 2, 3 and 4, we get

4 ,-wfx,)- (, -w)P
2 PA-"2 P =u -H =0

4 2 , -H fix)

In practice the first four moments suffice.

Mean deviation from mean
la, |fu,)
b)IfX is a continuous random variable with
probability density function flx), then
EX)= x fir) dx
EXX)= flx) dx

Var(X)- a-flr) ds EX*) -

IEX
Mean deviation from mean- x-fx)
ds
The moment about the mean is defined as
 mean are called central moments.
about the
Moments
Generating Function
(c) Moment
-a
function M,(t) = Pe's,
Consider the ..1)
This function is a function of the parameter t and gives the mean of the probabilitu

etribution of e-a), Expanding the exponential in equation (1), we oet

M0-21+tl,-a)2G;
2! -a* +..( -af +
-+ P. -a)2 P.-a. -a' +

2)

where u', is the moment of order r abouta. Thus M,(t) generates moments and is,
therefore,
called the moment generating function (m.g.f.) of the discrete probability distribution of the

variate X about the value x = a. We observe that u', is equal to the co-efficient of in the

expansion of m.g.f. M,t).

Alternately, u', can be obtained by differentiating (2) r times with respect to t and then
puttingt = 0.

Thus, K, =dM,(0)
Jt=0
Rewriting equation (1) as
M0) P,e o =e"ape=e"M,)
Hence the m.g.f. about the value a is eat times the m.g.f. about the orign.
AIS a continuous random variable with probability density function fx), then

M,)=-"fla)de
Note 1. If X is a random variable and a, b are constants, then

EaX+b) = aEX) +b
Var. (aX + b) = a* Var. (X) origin a

and moments about any arbitrary

telation between central moments

H'-3u',+2(',"
 ILLUSTRATIVE EXAMPLES

Example 1. Five defective bulbs are accidentally mixed with twenty good ones. It is not
possible to just look at a bulb and tell whether or not it is defective. Find the probability distri.
bution of the number of defective bulbs, if four bulbs are drauwn at random from this lot.
Sol. Let X denote the number of defective bulbs in 4. Clearly X can take the values 0, 1,
2, 3 or 4
Number of defective bulbs 5
Number of good bulbs 20
Total number of bulbs 25
PCX = 0) = P (no defective) = P (all 4 good ones)

C20 x 19x 18x 17 969

25C 25 x 24 x 23x 22 2530

PCX =1) =
P(one defective and 3 good 5CxCs 1140
ones)=25 C 2530

POX = 2) = P (2 defectives and 2 good ones) = C2x C2380

5 CA 2530

CxC- 40
PX = 3) = P(3 defectives and 1 good one) =

5C 1
CA 2530

PX 4) = P (all 4 defective):
2530
The probability distribution of the random variable X is
X 0 1 2 3
PCX) 969 1140 380 40
2530 2530
1
2530 2530 2530
Example 2. A die is tosse ne. A success is 'getting 1 or 6' ona toss. Find the mean
and the variance of the number of successes.
Sol. Let X denote the number of success.
Clearly X can take the values 0, 1, 2 or 3.
Probability of success =2- Probability of failure =1-

PCX = 0) = P (no success) = P (all 3 failures)

"27
PX =1) = P (one suecess and 2 failures) =
C, xxx
PCX = 2) = P (two successes and failure)
one =
C, xx 3
x

PX = 3) = P (all 3 successes) =
 The probability distribution of the random variable X is
X 1 2 3
8 12
PX) 6
27 27 27 27
To find the mean and variance

P
P PA
8
27
12
1
27 27 12
6 12
27 27
24
27
3 1 3
27 27
9
27

1 5
3
Mean
H=2p=1
Variance o =pa*-*
Example 3. A random variable X has the following probability function:
Values of X 1 2 3 4 5 6 7
p(x) 0 2k 2k 3 2 2k2 T+k
(i) Find k, (i) Evaluate PX <6), PX2 6), P3 <Xs6)
) Find the minimum value of r so that PCXSx)> 2.
Sol. (i) Since plx) = 1, we have
x =0

0+ + 2k + 2k + +2 +2k2 +7k2 +k =1
3k
10k2+9k -
1 = 0 (10k (10k - 1Xk+ 1) 00 =

:px) 0
k 1
10
(1) 1) . . + PMX
= 5)
PCX < 6) = P(X = 0) + PX =

81
= 0+k+2k + 2k +3k +k2 = 8k +k2=+ 100 100

PCX 26) =
PX = 6) + P(X =7)
19
2 2 + 7k2+k = 100 10 100
100
PX = 6)
=4) + PX 5) +
P3 <X s 6) = PX =

33
3 3100
3k +k2+2k= 1+ 100 100
 PCX S 2 ) = k + 2k = 1o

)=k=
(ii)PX S 1
+ 2k + 2k +
3k =

2k + 2k = P(X 4)k
PXS 3) =k +

is 4.
The maximum value of x so that P(X <1)>
tickets are draun at a time out of n tickets numbered fron
rom
Example 4. In a lottery, m numbers on the tickets draun.
n. Find the expected
value of the sum of the
1 to

Sol. Let X denote the numberon a ticket then X assumes vlaues 1, 2, 3,.. n

The probability of drawing a ticket out of n tickets is

n

Values of X, 1 2 3 7

P(X=«):
E (a) P, *
-1

2 +2+3+...... + 11)

nn+1)n+l
2 2
If the numbers on the m tickets drawn are , N9 n Where n; E{1, 2, 3,
i=1,2, ... m, then the expected value of the sum of numbers on m tickets
.....
,
1

= Er,) + E(u,) + ..+E ,

mln +1)
m ECx) =

2
Example 6. Is the function defined as follous a density function?
fa)= 20
f)=\0, otherwise
f so, fnd P(1 sXs 2).
Sol. fx) 20 for every x in (-o,
o)and

flo) d =|f)d+flx)da
-

0da+ ds =0+ =-(0-1) =1