SSCE 2193 Engineering Statistics Sem 2 1617

EXPONENTIAL DISTRIBUTION

The discussion of the Poisson distribution defined a random variable to be the number
of flaws along a length of copper wire. The distance between flaws is another random
variable that is often of interest. Let the random variable X denote the length from any
starting point on the wire until a flaw is detected.

As you might expect, the distribution of X can be obtained from knowledge of the
distribution of the number of flaws. The key to the relationship is the following concept.
The distance to the first flaw exceeds 3 millimeters if and only if there are no flaws within
a length of 3 millimeters - simple, but sufficient for an analysis of the distribution of X.

In general, let the random variable Y denote the number of flaws in x millimeters of
wire. If the mean number of flaws is λ per millimeter, Y has a Poisson distribution with
mean γx. We assume that the wire is longer than the value of x. Now
e(−γx) (γx)0
P (X > x) = P (N = 0) = = exp(−γx)
0!
Therefore
F (X) = P (X 6 x) = 1 − e(−γx) , x>0
is the cumulative distribution function of X. By differentiating F (X), the probability
density function of X is calculated to be
f (X) = γe(−γx) , x>0
The derivation of the distribution of X depends only on the assumption that the flaws
in the wire follow a Poisson process. Also, the starting point for measuring X doesn’t
matter because the probability of the number of flaws in an interval of a Poisson process
depends only on the length of the interval, not on the location. For any Poisson process,
the following general result applies.

Definition:

The random variable X that equals the distance between the successive counts of a
Poisson process with mean λ > 0 is an exponential random variable with parameter γ.
The probability density function of X is
f (X) = γe(−γx) , for 0 6 x < ∞
If the random variable X has an exponential distribution with parameter γ,
X ∼ Exp(γ)
1 1
µ = E(X) = and σ 2 = V ar(X) = 2
γ γ
 2

Example:

The time between phone calls received by a telephonist is exponentially distributed

with mean of 10 minutes.

a) What is the probability that there are no calls in one hour?

b) What is the probability that there are not more than four calls within one hour?

c) Determine x such that the probability that there are no calls within x hours is 0.02.

Solution:

Let the random variable X denote the length from any starting point on the time until
a phone call is received. As we might expect, the distribution of X can be obtained from
knowledge of the distribution of the number of calls. The key to the relationship is the
following concept. By converting all units to minutes, therefore we have the distance to
the first call exceeds 60 minutes (1 hour) if and only if there are no calls within a length
of 60 minutes.

In general, let the random variable Y denote the number of calls in x minutes of time.
If the mean number of calls is λ per minute, Y has a Poisson distribution with mean γx.
By assuming that the time is longer than the value of x, we have
1 1
a) E(X) = = 10 minutes ⇒ γ= per minute
γ  10
1
X ∼ Exp
10
1
P (X > 60) = P (Y = 0) = e(−γx) = e(− 10 (60))
= e(−6) = 0.0025
 
1
b) Y ∼ Po(γx) ∼ Po (60) ∼ Po(6)
10

P (Y 6 4) = 0.2851 (From Statistical Table)

c) The distance to the first call exceeds x minutes if and only if there are no calls
within a length of x minutes

P (X > x) = P (Y = 0) = e(−γx)
1
0.02 = e(− 10 (x))
x
− = ln(0.02) = −3.912
10
x = 39.12 minutes
 3

Additional Example:

In a large corporate computer network, user log-ons to the system can be modeled as
a Poisson process with a mean of 25 log-ons per hour.

a) What is the probability that there are no log-ons in an interval of 6 minutes?

b) What is the probability that the time until the next log-on is between 2 and 3
minutes?

c) Determine the interval of time such that the probability that no log-on occurs in
the interval is 0.90.

d) Find the mean and standard deviation of the time until the next log-on.

Solution:

Let X denote the time in hours from the start of the interval until the first log-on.
Then X has an exponential distribution with λ = γx = 25 log-ons per hour. We are
interested in the probability that X exceeds 6 minutes. Because λ is given in log-ons per
6
hour, we express all time units in hours. That is, 6 minutes = 60 = 0.1 hour. Y denotes
as the number of log-ons, is having Poisson distribution with mean γx.

a) λ = γx = 25 per hour ⇒ x = 1 hour then γ = 25 ⇒ X ∼ Exp(25)

P (X > 0.1) = P (Y = 0) = e(−γx) = e(−25(0.1))

= e(−2.5) = 0.0821

b) Upon converting all units to hours:

2 1 3 1
2 minutes = 60 = 30 hour and 3 minutes = 60
= 20
hour.
  Z 1
1 1 20
P <X< = 25e(−25x) dx
30 20 1
30
 (−25x)  201
= −e 1
30
h 5
i h 5
i
= −e(− 4 ) − −e(− 6 ) = 0.1481

c) The question ask for the length of time X such that P (X > x) = 0.90

P (X > x) = P (Y = 0) = e(−γx)
0.90 = e(−25x)
−25x = ln(0.90) = −0.1054
x = 0.004216 hour = 0.2530 minute
 4

1 1
d) The mean of the time until the next log-on is µ = E(X) = = = 0.04 hour =
γ 25
2.4 minutes The standard deviation of the time until the next log-on is

1 1 1
σ 2 = V ar(X) = = = = 0.0016 hour2 = 0.096 minutes2
γ2 252 625
p √
σ = V ar(X) = 0.0016 hour2 = 0.04 hour = 2.4 minutes

A LACK OF MEMORY PROPERTY:

An important property of the exponential distribution is that it is memoryless, which

means that if a random variable X is exponentially distributed, its conditional probability
is given by
P (X > x1 + x2 |X > x1 ) = P (X > x2 ) all x1 , x2 > 0
P (X < x1 + x2 |X > x1 ) = P (X < x2 )

Example:

The number of hits on a website follows a Poisson process with rate of four per minute.

a) What is the probability that more than two minutes go without a hit?

b) If two minutes have gone by without a hit, what is the probability that a hit will
occur in the last next minute?

Solution:

a) λ = γx = 4 per minute (x = 1min ⇒ γ = 4) ⇒ X ∼ Exp(4)

P (X > 2) = P (Y = 0) = e(−γx) = e(−4(2))

= e(−8) = 3.3546 × 10(−3)

b) x1 = 2, x2 = 1

P (X < x1 + x2 |X > x1 ) = P (X < x2 )

P (X < 2 + 1|X > 2) = P (X < 3|X > 2) = P (X < 1)
= 1 − e(−λx) = 1 − e(−4(1)) = 0.9817
 5

Additional Example:

Let X denote the time between detections of a particles with a gieger counter and
assume that X has an exponential distribution with mean 1.4 minutes.

a) What is the probability that a particle can be detected within 30 seconds of starting
the counter?

b) Suppose the gieger counter has been turn on and wait 3 minutes without detecting a
particle. What is the probability that a particle is detected in the next 30 seconds?

Solution:

1 1
E(X) = = 1.4 minutes ⇒ γ= per minute ⇒ X ∼ Exp(1.4)
γ 1.4

a) The probability that a particle can be detected within 30 seconds of starting the
counter is
30
60
= 0.5 minute (All units converted to minutes)

1
P (X < 0.5) = 1 − e(−λx) = 1 − e(− 1.4 (0.5)) = 0.3003

b) The probability that a particle is detected in the next 30 seconds after the gieger
counter has been turn on and wait 3 minutes without detecting a particle is

P (X < x1 + x2 |X > x1 ) = P (X < x2 )

P (X < 3 + 0.5|X > 3) = P (X < 3.5|X > 3) = P (X < 0.5)
1
= 1 − e(−λx) = 1 − e(− 1.4 (0.5)) = 0.3003

0R

P (3 < X < 3.5)

P (X < x1 + x2 |X > x1 ) =
P (X > 3)
P (3 < X < 3.5) = F (3.5) − F (3)
h 1
i h 1
i
= 1 − e(− 1.4 (3.5)) − 1 − e(− 1.4 (3)) = 0.0352

1
P (X > 3) = e(−λx) = e(− 1.4 (3)) = 0.1173

P (3 < X < 3.5)

P (X < 3.5|X > 3) =
P (X > 3)
0.0352
= ≈ 0.3001
0.1173