Exponential Distribution
Exponential Distribution Definitions
The exponential distribution is a continuous distribution that is
commonly used to measure the expected time for an event to occur.
OR
It can also be used to measure the likelihood of incurring a specified
number of defaults within a specified time period.
OR
The exponential distribution has been successfully applied as a time-
to-failure model for complex systems consisting of a large number of
components in series, none of which individually contributes
significantly to the total failure density.
Real Life Applications
The exponential distribution has many application in real life.
Some of the application are given below:
1. The amount of time (beginning now) until an earthquake occurs has
an exponential distribution.
2. The length, in minutes, of long distance business telephone calls.
3. The amount of time, in months, a car battery lasts.
4. In physics it is often used to measure radioactive decay.
5. In engineering it is used to measure the time associated with
receiving a defective part on an assembly line.
6. In finance it is often used to measure the likelihood of the next
default for a portfolio of financial assets. It can also be used to
measure the
Probability Density Function
A random variable X is said to have an exponential distribution with a
parameter or rate parameter > 0 if it follows the following pdf.
𝑓 𝑥 = 𝜆𝑒 𝑓𝑜𝑟 0 ≤ 𝑥 ≤ ∞
0 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒
Derivation of the pdf
Let X denote the waiting time until the first change occurs. Then X is a
continuous random variable and we determine its density function.
𝐹 𝑥 = 𝑃 𝑋 ≤ 𝑥 = 1 − 𝑃(𝑋 > 𝑥)
It means if you have to wait for at least x units for the first event to occurs that
means there is no change occurs from [0,x] , it can be written as
𝐹 𝑥 = 1 − 𝑃 𝑋 > 𝑥 = 1 − 𝑃(𝑛𝑜 change in 0, x )
And that turns out to be a poisson random variable with some mean. If the mean
number of event in interval of length 1 is 𝜆 then mean number of event of the length
x is 𝜆x (mean waiting time for first change to occur) i.e.
𝐹 𝑥 = 1 − 𝑃 𝑛𝑜 events in 0, x = 1 − f(0; 𝜆𝑥)
𝑒 𝜆𝑥
𝐹 𝑥 = 1−
0!
𝐹 𝑥 =𝑃 𝑋 ≤𝑥 =1−𝑒
The cdf of exponential distribution can be written as
𝐹 𝑥 = 1−𝑒 𝑥>0
0 𝑥≤0
Now by differentiating F(x) with respect to x, we get the probability density
function of X as
𝑓 𝑥 = 𝜆𝑒 𝑥>0
0 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒
Mean and variance of exponential distribution
𝑒 1
𝜇= 𝑦 𝑓 𝑦 𝑑𝑦 = 𝑦𝜈𝑒 𝑑𝑦 = −𝑦𝑒 + 𝑒 𝑑𝑦 = − =
𝜈 𝜈
1
𝜎 = 𝑦 𝑓 𝑦 𝑑𝑦 − 𝜇 = 𝑦 𝜈𝑒 𝑑𝑦 −
𝜈
1 𝜇 1 1
= −𝑦 𝑒 +2 𝑦𝑒 𝑑𝑦 − =0+2 − =
𝜈 𝜈 𝜈 𝜈
𝜎 𝜎
𝜈=3
1
𝜇=
3
Questions
An average of 0.5 ships per day arrives at certain dock. What is the
probability that after the departure of a ship four days will pass before the arrival of
the next ship?
Solution
We are given with average per day i.e. 𝜆 = 0.5 ships. So
∞
𝑃 𝑋 > 4 = 𝑓 𝑥; 𝜆 𝑑𝑥 = 𝜆𝑒 𝑑𝑥 = −𝑒
4
∞
𝑃 𝑋 > 4 = −𝑒 . =− 0−𝑒 =𝑒
4
P(X > 4) = 0.13534
Or it can also be solved as
𝑃 𝑋 >4 =1−𝑃 𝑋 ≤4
From the cdf of exponential distribution we know that
𝐹 𝑥 = P X≤𝑥 = 1−𝑒
So
𝑃 𝑋 >4 =1−𝑃 𝑋 ≤4 =1− 1−𝑒 . = 𝑒 = 0.13534
Example: Reliability
The time till failure of an electronic component has an Exponential distribution and it
is known that 10% of components have failed by 1000 hours.
(a) What is the probability that a component is still working after 5000 hours?
(b) Find the mean and standard deviation of the time till failure.
Answer
Let Y = time till failure in hours; 𝑓 𝑦 = 𝜈𝑒 .
(a) First we need to find 𝜈 𝑃 𝑌 ≤ 1000 = 𝜈𝑒
= −𝑒 =1−𝑒
𝑃 𝑌 ≤ 1000 = 0.1 ⇒ 1−𝑒 = 0.1
⇒𝑒 = 0.9
⇒ −1000𝜈 = ln 0.9 = −0.10536 ⇒ 𝜈 ≈ 1.05 × 10
If 𝑌 is the time till failure, the question asks for 𝑃(𝑌 > 5000):
𝑃 𝑌 > 5000 = 𝜈𝑒 𝑑𝑦
= −𝑒 =𝑒 ≈ 0.59
(b) Find the mean and standard deviation of the time till failure.
Answer: Mean = 1/𝜈 = 9491 hours.
Standard deviation = Variance
= = 1/𝜈 = 9491 hours
Practice Questions
Q 1. An average of 0.5 customers per minute arrives at a restaurant. After an
attendant opens the restaurant, what is the probability that there will be a wait of at
least 3 minutes before the first customer arrives?
Q 2. The duration of long distance telephone calls is found to be exponentially
distributed with a mean of 3 minutes. What is the probability that a call will last
i) more than 3 minutes ii) between 3 to 5 minutes
Q 3. Let be an exponential random variable with parameter . Compute the
following probability.