In many random experiments, we are interested not in knowing which of the outcomes has occurred

but in the numbers associated with them. For example, when n coins are tossed, one may be
interested in knowing the number of heads obtained. When a pair of dice are tossed, one may seek
information about the sum of points. Thus, we associate a real number with each outcome of a
random experiment. In other words, we are considering a function whose domain is the set of all
possible outcomes and whose range is a subset of the set of reals.
Definition Let S be the sample space associated with a given random experiment. A real-valued
function X: S→ (−∞, ∞) is called a one-dimensional random variable(r.v.).
Consider the random experiment of tossing a coin. Then S={w1,w2}, w1=H,w2=T. Define
X:{w1,w2}→{0,1} by X(w1)=1, X(w2)=0. X is a r.v.
A function X:S→R2 is a two-dimensional random variable.
: If a dart is thrown at a circular target, the sample space S is the set of all points w on
the [Link] imagining a co-ordinate system placed on the target with the origin at the centre, we
can consider a two-dimensional random variable X which assigns to every point w of the circular
region , its rectangular co-ordinates (x,y)
If a pair of dice is tossed , then S={1,2,3,4,5,6}X{1,2,3,4,5,6} .Let X be the random
variable defined by X(i,j)=max{i,j}. Then
P(X=1)=P{(i,j):X(i,j)=1}=P{(1,1)}=1/36, P(X=2)=P{(1,2),(2,2),(2,1)}=3/36.

Distribution Function
Definition Let X be a random variable(r.v.). The function F : (-∞,∞)→[0,1] defined by F(x)=P{t:
X(t)≤x} is the distribution function (d.f.) of the r.v. X.
Note: To emphasize the r.v. X, we sometimes denote F(x) by FX(x).

Discrete Random Variable

A r.v. which can assume only at most countable number of real values is a discrete random variable.
Example of discrete random variable are marks obtained in a test, number of accidents per month
etc.
Probability Mass Function
If X is a one-dimensional discrete r.v. taking at most a countable number of values x1,x2,…, then the
probabilistic behaviour of X at each xi is described by its probability mass function.

1
Definition If X is a discrete r.v. having distinct values x1,x2,…, then the function pX(x), or simply p(x),
defined by p(x)=P(X=xi)=pi, if x=xi and =0, if x≠xi, i=1,2,… is called probability mass function(p.m.f.)
of r.v. X.
Note (1)The set {(x1,p1),(x2,p2),…} specifies the probability distribution of the r.v. X.
(2) P(xi)≥0 , for all i and ∑∞
1 𝑝𝑝(𝑥𝑥𝑖𝑖 ) = 1.

Let S={H,T} be the sample space corresponding to the random experiment of tossing
of a ‘fair’ coin. Let X be the r.v. defined by X(H)=1, X(T)=0. X has only two distinct values, namely, 0
and 1. The corresponding p.m.f. is given by: p(1)=P(X=1)=P(H)=1/2, p(0)=1/2.
A r.v. X has the following p.m.f.:
xi: 0 1 2 3 4 5 6 7
pi: 0 k 2k 2k 3k k2 2k2 7k2+k
(1)Find k, (2)Evaluate P(X<6),P(X≥6) and P(0<X<5), (3) If P(X≤a)>1/2, find the minimum value of
a, (4) determine the p.d.f. of X.
Solution: (1) Since , k+2k+2k+3k+k2+2k2+7k2+k=1 giving 10k2+9k-1=0, which gives k=1/10 or
-1. Since p2=k cannot be negative, -1 is rejected and k=1/10.

(2)P(X<6)=P(X=0)+P(X=1)+…+P(X=5)=1/10+2/10+2/10+3/10+1/100=81/100.

Now P(X≥6)=1-P(X<6)=1-81/100=19/100.

(1) P(X≤a)>1/2. By trial, we get a=4.

(2) The p.d.f. of X is given by:

X: 0 1 2 3
F(x)=P(X≤x): 0 k=1/10 3k=3/10 5k=5/10

X: 4 5 6 7
F(x) 8k=4/5 8k+k 8k+3k 9k+10k2
2 2

If p(x)=x/15, x=1,2,3,4,5; =0, elsewhere be the p.m.f. of a r.v.X. Find (1) P{X=1 or 2},
1 5
(2) P{2 < 𝑋𝑋 < 2 | 𝑋𝑋 > 1}.

 (1) P{X=1 or 2}=P(X=1)+P(X=2)=1/15+2/15=1/5.

1 5 2
1 5 𝑃𝑃{� <𝑋𝑋< �∩(𝑋𝑋>1)} 𝑃𝑃((𝑋𝑋=1 𝑜𝑜𝑜𝑜2)∩(𝑋𝑋>1)) 𝑃𝑃(𝑋𝑋=2) 1
(1) P{2 < 𝑋𝑋 < 2 | 𝑋𝑋 > 1} = 2 2
= = 1−𝑃𝑃(𝑋𝑋=1) = 15
1 = 7.
𝑃𝑃(𝑋𝑋>1) 𝑃𝑃(𝑋𝑋>1) 1−
15

2
Continuous Random Variable
Definition A r.v. X is continuous iff X takes all values between two unequal real numbers.
Probability Density Function
Consider a small interval (x, x+dx) of length dx about x. Let f(x) be any continuous function of x so
that f(x)dx represents the probability that X falls in the infinitesimal interval (x, x+dx). Symbolically,
P(x≤ 𝑋𝑋 ≤ 𝑥𝑥 + 𝑑𝑑𝑑𝑑)=fX(x)dx. fX is called probability density function (p.d.f.) of the r.v. X.
𝑏𝑏
The probability for a variate value to lie in the interval [a,b] is P(a≤X≤b)=∫𝑎𝑎 𝑓𝑓𝑋𝑋 (𝑥𝑥)𝑑𝑑𝑑𝑑.

The p.d.f. f(x) of a r.v. X has the properties:

∞ ∞
• f(x)≥0, ∫−∞ 𝑓𝑓(𝑥𝑥)𝑑𝑑𝑑𝑑=1 (since ∫−∞ 𝑓𝑓(𝑥𝑥)𝑑𝑑𝑑𝑑 gives total probability),
𝑐𝑐
• P(X=c)= ∫𝑐𝑐 𝑓𝑓𝑋𝑋 (𝑥𝑥)𝑑𝑑𝑑𝑑=0 (where c is any value of the variate X)

The diameter of an electric cable , say X, is assumed to be a continuous random

variable with p.d.f. f(x)=6x(1-x), 0≤x≤1. (1) Check that f(x) is a p.d.f., (2) determine the median b of
the distribution.
1
 (1) ∫0 𝑓𝑓(𝑥𝑥)𝑑𝑑𝑑𝑑 = 1 (by direct calculation); hence f(x) is p.d.f. of r.v. X. (2)
𝑏𝑏 1
P(X<b)=P(X>b)⇒6 ∫0 𝑓𝑓(𝑥𝑥)𝑑𝑑𝑑𝑑 = 6 ∫𝑏𝑏 𝑓𝑓(𝑥𝑥)𝑑𝑑𝑑𝑑⇒ b=1/2, lying in [0,1].

Suppose that the life in hours of a certain part of radio tube is a continuous random
variable X with p.d.f. given by f(x)=100/x2, when x≥100; =0, elsewhere. (1) What is the probability
that all of three such tubes in a given radio set will have to be replaced during the first 150 hours of
operation?(2)What is the probability that none of the original tubes will have to be replaced during
the first 150 hours of operation?(3)what is the probability that a tube will last less than 200 hours
if it is known that the tube is still functioning after 150 hours of service?
150 150 100 1
 (1)p=P(X ≤ 150)= ∫100 𝑓𝑓(𝑥𝑥)𝑑𝑑𝑑𝑑 = ∫100 𝑥𝑥 2
𝑑𝑑𝑑𝑑 = 3 . By compound probability theorem, the
probability that all three of original tubes will have to be replaced during the first 150 hours
=p3=1/27.
(2)The probability that a tube is not replaced during the first 150 hours of operation is
P(X>150)=1-P(X≤150)=1-p=2/3. By compound probability theorem, the probability that
none of the three tubes will have to be replaced during the first 150 hours =q3=8/27.
(3)Probability that a tube will last less than 200 hours given that the tube is still functioning
200100
𝑃𝑃(150<𝑋𝑋<200) ∫ 𝑑𝑑𝑑𝑑 1 3
𝑥𝑥2
after 150 hours is P(X<200|X>150)= 𝑃𝑃(𝑋𝑋>150) = 150∞ 100 = 6 𝑋𝑋 2=0.25.
∫150 2 𝑑𝑑𝑑𝑑
𝑥𝑥

The amount of bread (in hundreds of pounds) x that a certain bakery is able to sell
in a day is found to be a numerical valued random phenomenon with a probability function specified
by the p.d.f. f(x) given by f(x)=kx, 0≤x<5; =k(10-x), 5≤x<10; =0, otherwise. (1) find the value of k
such that f(x) is a p.d.f., (2)what is the probability that the number of pounds of bread that will be

3
sold tomorrow is (a) more than 500 pounds, (b) less than 500 pounds and (c) between 250 and 750
pounds? (3) Denoting by A,B,C the events that the pounds of bread sold are as in (2)(a),(2)(b) and
(2)(c) respectively, find P(A|B),P(A|C). Are (1) A,B independent, (2) A,C independent?
∞
 (1)∫−∞ 𝑓𝑓(𝑥𝑥)𝑑𝑑𝑑𝑑=1 gives k=1/25.
10 1
(2)(a)P(5≤ 𝑋𝑋 ≤10)=∫5 (10 − 𝑥𝑥)𝑑𝑑𝑑𝑑=0.5
25
5 1
(b)P(0≤ 𝑋𝑋 < 5) = ∫0 25 𝑥𝑥𝑥𝑥𝑥𝑥 =0.5
5 1 7.5 1
(c)P(2.5≤ 𝑋𝑋 ≤ 7.5) = ∫2.5 25 𝑥𝑥 𝑑𝑑𝑑𝑑 + ∫5 25 (10 − 𝑥𝑥)𝑑𝑑𝑑𝑑=3/4
(3)From (2)(a),(b),(c), P(A)=0.5, P(B)=0.5, P(C)=3/4. The events A∩B and A∩ 𝐶𝐶 are given by:
1 7.5 1
A∩B=∅, A∩ 𝐶𝐶:5<X<7.5. Thus P(A∩B)=0,P(A∩C)= ∫5 (10 − 𝑥𝑥)𝑑𝑑𝑑𝑑=3/8.
25 25
P(A)P(C)=1/2 X ¾=3/8=P(A∩ 𝐶𝐶), P(A)P(B)=1/4≠ 0 =, 𝐵𝐵 𝑎𝑎𝑎𝑎𝑎𝑎 𝑛𝑛𝑛𝑛𝑛𝑛 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 P(A∩B). Thus
A,C are independent and A,B are not independent.

Expectation of a r.v.
Let the r.v. X take values x1,…,xn with probabilities p1,…,pn. Let X take value xi , fi number of times;
𝑓𝑓1 𝑥𝑥1 +⋯+𝑓𝑓𝑛𝑛 𝑥𝑥𝑛𝑛 𝑓𝑓1 𝑓𝑓
let N=f1+…+fn. Mean of X is given by 𝑥𝑥 + ⋯ + 𝑁𝑁𝑛𝑛 𝑥𝑥𝑛𝑛 . Let N→ ∞. Using the statistical
=
𝑁𝑁 𝑁𝑁 1
definition of probability, limiting value of mean of X , 𝑋𝑋� = 𝑝𝑝1 𝑥𝑥1 + ⋯ + 𝑝𝑝𝑛𝑛 𝑥𝑥𝑛𝑛 .
Definition Expectation of X, E(X)=∑𝑛𝑛1 𝑝𝑝𝑖𝑖 𝑥𝑥𝑖𝑖 .
Thus E(X) may be regarded as the limiting value of the average value of X realized in N random
experiments as N→ ∞. Generalising, if f(X) is a function of X, f(X) will take values f(x1),…,f(xn) with
𝑓𝑓 𝑓𝑓
frequencies f1,…,fn and the average value of f(X) in N experiments is 𝑁𝑁1f(x1)+…+ 𝑁𝑁𝑛𝑛f(xn) and as N→ ∞,
/
this approaches to E(f(X))=p1f(x1)+…+pnf(xn). In particular, 𝜇𝜇𝑟𝑟 (𝑎𝑎)=E[(X-a)r],𝜇𝜇𝑟𝑟 = 𝐸𝐸[(𝑋𝑋 − 𝑋𝑋�)r], 𝜇𝜇2 =
𝐸𝐸[(𝑋𝑋 − 𝑋𝑋�)2]=𝜎𝜎 2 .
What is the expectation of the number of failures preceding the first success in an
infinite series of independent trials with constant probability p of success in a trial?
 If X denotes the number of failures preceding the first success , we find that X takes the values
0,1,2,3,… with probabilities p,qp,q2p,q3p,…, where q=1-p. Thus probability density function
is f(x)=qrp, r=0,1,2,…. Hence E(X)=0.p+[Link]+2.q2p+3.q3p+…=pq(1+2q+3q2+…)=pq(1-q)-2
(since q<1)=q/p = 1/p-1.

Properties of Expectation
1. Addition Theorem of Expectation: If X,Y are r.v., then E(X+Y)=E(X)+E(Y).
2. Multiplication Theorem of Expectation: If X,Y are independent r.v., E(XY)=E(X)E(Y).
3. If X is a r.v. and a,b are constants, then E(aX+b)=aE(X)+b, provided all the expectations exist.
4. If X≥0,then E(X) ≥0.
5. If X,Y are r.v. and X(t) ≥Y(t), forall t, then E(X) ≥E(Y), provided all expectations exist.

4
 Let X be a r.v. with the following probability distribution:
x: -3 6 9
P(X=x): 1/6 ½ 1/3
Find E(X) and E(X2) and using laws of expectation, evaluate E(2X+1)2.
1 1 1 11 93
 E(X)=∑ 𝑥𝑥𝑥𝑥(𝑥𝑥)=(-3)6 + 6 2 + 9 3= 2 , E(X2)=∑ 𝑥𝑥 2 𝑝𝑝(𝑥𝑥) = 2
. Then
E(2X+1)2=4E(X2)+4E(X)+1=209.
Two unbiased dice are thrown. Find the expected values of the sum of numbers of
points on them.
 The probability function of X(sum of number of heads on two dice) is
x: 2 3 4 … 12
P(X=x): 1/36 2/36 3/36 … 1/36
1 1
E(X)=∑ 𝑝𝑝𝑝𝑝 = 36 (2 + 6 + 12 + 20 + 30 + 42 + 40 + 36 + 30 + 22 + 12) = 36 252=7.