Professional Documents
Culture Documents
x > 0.
Recall that the function by which we get the cumulative probability P (X < x) is called the
cumulative density function (CDF) or simply, the distribution function (DF), and is denoted
1
0.00
0.01
0.02
f(x)
0.03
0.04
0.05
Exponential Distribution
20
40
60
80
100
xaxis
x > 0.
Solution: To solve this problem, we shall convert the service rate so that the time period is
1 minute. So, as the service rate is 6 per hour, we shall take the parameter to be 0.1 per
minute.
(a) P (X < 5) = 1 ex = 1 e0.1(5) = 0.3935
(b) P (X > 10) = ex = e0.1(10) = 0.3679.
The mean of exponential distribution is 1/, and the variance is 1/2 .
The exponential distribution is widely used to model lifetime data, i.e., data regarding lives
of machines, or of electrical appliances etc. The histograms of such data often look like Figure
2. This should explain why exponential distribution is a good choice to model this type of data.
The exponential distribution has an interesting property, called the memoryless property.
If X is an exponential random variable, then this property says
This property means, from the point of view of interarrival times, that it does not matter how
long you have waited so far in a queue for a service. It means, for example, that if you wait in
a queue for 10 minutes, the probability that you have to wait for another 10 minutes in that
queue does not depend on the information that you have already waited for 10 minutes.
From the view point of modeling lifetime data, this property does not always seem to be
very practical, and thus, it restricts the use of exponential distribution for such type of data
sometimes.
3
Figure 2: Histogram
2000
1000
0
Frequency
3000
4000
20
40
60
x
80
100