MATH F113

(Probability and Statistics)

Chandra Shekhar
Associate Professor

Department of Mathematics
BITS Pilani, Pilani Campus, Rajasthan 333 031
Email: chandrashekhar@pilani.bits-pilani.ac.in
Mobile: 9414492349
Chandra Shekhar MATH F113 (Probability and Statistics
What have you covered?

In Lecture 11
Application of Discrete Distribution
Binomial Approximation
Poisson Approximation

Chandra Shekhar MATH F113 (Probability and Statistics

 Chapter 4
Continuous Distribution

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Continuous Densities
Consider the random variable T ,
the time of the peak demand for
electricity at a particular power
plant.
Here, we cannot limit the set of
possible values for T to some
finite or countable infinite
collection of times.
Chandra Shekhar MATH F113 (Probability and Statistics
Continuous Distribution (Cont...)

Continuous Densities
Time T can conceivably assume
any value in the time interval
[0, 24), where 0 denotes mid night
of one day and 24 denotes 12 mid
night of next day.

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Continuous Densities
Further, we could pose the
question, what is the probability
that the peak demand will occur
exactly 12.013278. . . ?
It is virtually impossible for the
peak load to occur at this split.
Hence the answer is zero.

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Continuous Densities
Suppose the range of X is made
up of large finite number of
values say, X in 0 ≤ x ≤ 1 of the
form 0, 0.01, 0.02,. . . ,0.99, 1.0.
Each of the values is associated
with a non-negative number
whose sum is 1.
X can assume all possible values
0 ≤ x ≤ 1.
Chandra Shekhar MATH F113 (Probability and Statistics
Continuous Distribution (Cont...)

Continuous Densities
Since possible values of X are
non countable in 0 ≤ x ≤ 1, then
what happens to point
probabilities? We can’t really
speak of ith value of X for all X,
P [X = xi] becomes meaningless

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Definition
Continuous Random Variable
A random variable is continuous if
it can assume any value in some
interval (or intervals) of real
numbers and the probability that it
assume any specific value is 0
(zero).

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Definition
Continuous Density
Let X be a continuous random variable. A
function f (x) is called continuous density
(probability density function i.e pdf ) if
(i) fR(x) ≥ 0
+∞
(ii) −∞ f (x)dx = 1
(iii) For any a, b (real) with −∞ < x < +∞
we have
Z b
P [a ≤ x ≤ b] = f (x)dx
a
Chandra Shekhar MATH F113 (Probability and Statistics
Continuous Distribution (Cont...)

Remark 1: If f (x) is not pdf (or den-

sity) of X if
Z b
f (x)dx = k,
a
f (x)
k is not one, then k is the pdf of X.

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Remark 2: X is continuous r.v, X

assumes all values in (a, b), where a, b
may be replaced by −∞ and +∞ re-
spectively. We are considering the
idealized description of X.

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Remark 3: It is a consequence of the

probabilistic description that for any
specified value of X, say x0, we have
P [X = x0] = 0, since
Z x0
P [X = x0] = f (x)dx = 0
x0

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Remark 4: However, if you allow X

to assume all values in some interval,
then probability zero is not equiv-
alent with impossibility. Hence for
continuous case P (A) = 0 does not
imply A = Φ(empty set)

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Remark 5: If X assumes values in

some finite interval [a, b], we simply
set f (x) = 0 for all x 6∈ [a, b].

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Remark 6: Consider the line segment

{X | 0 ≤ x ≤ 2}
Every conceivable point on the line
segment could be the outcome of the
experiment.

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Since X is continuous r.v, we have

P [a ≤ x ≤ b]
= P [a ≤ x < b]
= P [a < x ≤ b]
= P [a < x < b]

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Example: Suppose that the r.v X

is continuous, Let the pdf f is given
by:
(
2X 0 < x <1
f (x) =
0, elsewhere
Clearly, f (x) ≥ 0 and also
Z∞ Z1
f (x)dx = 2xdx = 1
−∞ 0
Chandra Shekhar MATH F113 (Probability and Statistics
Continuous Distribution (Cont...)

To compute
Z1/2
1
P [X ≤ 1/2] = 2xdx =
4
0

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Conditional Probability: For exam-

ple,
P [X ≤ 1/2|1/3 ≤ X ≤ 2/3],
P [1/3 ≤ X ≤ 1/2]
=
P [1/3 ≤ X ≤ 2/3]
1/2
R
2xdx
1/3 5
= =
2/3
R 12
2xdx
1/3
Chandra Shekhar MATH F113 (Probability and Statistics
Continuous Distribution (Cont...)

Definition
Let X be the continuous r.v. with density
f (x). The cumulative distribution
function (cdf ) for X, denoted by
F (X), is defined by
F (X) = P (X ≤ x) ∀ x
Zx
= f (s)ds
−∞
Chandra Shekhar MATH F113 (Probability and Statistics
Continuous Distribution (Cont...)

PROBABILITY by using cdf F (x)

P (a ≤ X ≤ b) = F (b) − F (a).

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

For example, in the previous case,

we have
(
2x 0<x<1
f (x) =
0 elsewhere
Then, we can compute F (x).

0R if x < 0


x
F (x) = 0 2sds = x2 if 0 ≤ x < 1

1 if x ≥ 1


Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Theorem
Let F be the continuous cdf of a
continuous r.v with pdf f , then
d
f (x) = F (X)
dx
for all x at which F is differential

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Exercise 1/4.1/pp 138

Consider the function
f (x) = k x 2≤x≤4
(a) Find the value of k that makes
this a density for a continuous ran-
dom variable.

Z4
1
kxdx = 1 =⇒ k =
6
2
Chandra Shekhar MATH F113 (Probability and Statistics
Continuous Distribution (Cont...)

(b) Find P [2.5 ≤ X ≤ 3]

Z3
1
P [2.5 ≤ x ≤ 3] = xdx = 0.2292
6
2.5

(c) Find P [X = 2.5]

Z2.5
1
P [x = 2.5] = xdx = 0
6
2.5

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

(d) Find P [2.5 < X ≤ 3]

P [2.5 < x ≤ 3] = P [2.5 ≤ x ≤ 3] = 0.2292

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Exercise 3/4.1/pp. 139

Let X denote the length in minutes
of a long distance telephone conver-
sation. Assume that density for X is
given by
1 −x
f (x) = e 10 , x>0
10

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Exercise 3/4.1/pp. 139

(a) Verify that f is a density for a
continuous random variable.
x
f (x) > 0, since for all x, e− 10 > 0 and
Z∞ Z0 Z∞
f (x)dx = f (x)dx + f (x)dx = 1
−∞ −∞ 0

Hence, f (x) is a pdf

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

(b) Assume that f adequately de-

scribes the behavior of the random
variable X, find the probability that
a randomly selected call will last at
most 7 minutes; at least 7 minutes;
exactly 7 minutes.

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

At most 7 minutes, we have

Z0 Z7
P [X ≤ 7] = f (x)dx + f (x)dx
−∞ 0

Z7
1 −x
=0+ e 10 dx = 0.5034
10
0

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Exactly 7 minutes, we have

P [X = 7] = 0
Probability of at least 7 minutes, we
have
P [x ≥ 7] = 1−P [x ≤ 7] = 1−0.5034 = 0.4966

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

(c) Would it be unusual for a call to

last between 1 and 2 minutes? Ex-
plain, based on the probability of this
occurring.
We have
Z2
1 −x
P [1 < x < 2] = e 10 dx = 0.09861
10
1

Which is relatively small value

Chandra Shekhar MATH F113 (Probability and Statistics
Continuous Distribution (Cont...)

(d) What is c.d.f for the above p.d.f ?

Given that
1 x
f (x) = e− 10 , x > 0
10
Zx
1 −t
F (x) = e 10 dt
10
0
x
= 1 − e− 10 , x > 0

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Hence
(
0 x<0
F (x) = x
1 − e− 10 x≥0

Chandra Shekhar MATH F113 (Probability and Statistics

Continuous Distribution (Cont...)

Exercise: Find the CDF of the fol-

lowing density function
(i)

1/3 0 ≤ x ≤ 1

f (x) = 2/3 1 < x ≤ 2

0 elsewhere

(ii)
(
|x| −1 < x < 1
f (x) =
0 elsewhere
Chandra Shekhar MATH F113 (Probability and Statistics