Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Continuous Probability Distributions

Introduction
Random variables may be classified in two distinct categories called discrete random variables and continuous
random variables. Discrete random variables can take values which are discrete and which can be written in the
form of a list. In contrast, continuous random variables can take values within a continuous range or an interval,
like the temperature in Central Park, or the height of an athlete in centimeters.
The most commonly met continuous random variables in engineering are the Uniform, Exponential and Normal
distributions.

Probability Density Function (PDF)


Although the probability distribution of the continuous random variable 𝑿 cannot be presented in tabular form,
it can be represented by a formula. This formula will be a function of the numerical values of the continuous
variable 𝑿 and will be denoted by 𝒇(𝒙), which is usually called the probability density function of random
variable X (pdf).

Definition: The probability distribution or the probability density function (pdf) of a continuous random
variable 𝑿 is a function 𝒇(𝒙) defined on some interval 𝒂 ≤ 𝑿 ≤ 𝒃 such that:

𝒃
𝑷(𝒂 ≤ 𝑿 ≤ 𝒃) = ∫𝒂 𝒇(𝒙) 𝒅𝒙
The probability density function is constructed so that the area under its curve bounded by the x- axis is equal to
1 when computed over the entire range of X when 𝒇(𝒙) is defined.

𝑷(𝒂 ≤ 𝑿 ≤ 𝒃) = ∫ 𝒇(𝒙) 𝒅𝒙
𝒂

Finding the normalization constant:


Not every real-valued function can represent a pdf density, it must satisfy the following two conditions:
1- 𝒇(𝒙) ≥ 𝟎 𝒇𝒐𝒓 𝒂𝒍𝒍 𝒙 ∈ ℝ

2- ∫−∞ 𝒇(𝒙)𝒅𝒙 = 𝟏 𝒂𝒓𝒆𝒂 𝒖𝒏𝒅𝒆𝒓 𝒕𝒉𝒆 𝒆𝒏𝒕𝒊𝒓𝒆 𝒈𝒓𝒂𝒑𝒉 𝒐𝒇 𝒇(𝒙).

Definition: Given an arbitrary positive function, 𝒇(𝒙) ≥ 𝟎, defined on some interval 𝒂 ≤ 𝑿 ≤ 𝒃 then the
𝒃
process of choosing a suitable constant 𝒌 to make ∫𝒂 𝒌. 𝒇(𝒙) 𝒅𝒙 = 𝟏 is called normalizing the function 𝒇(𝒙).
In other words, if we are asked to normalize the function 𝒌. 𝒇(𝒙), on an interval [𝑎, 𝑏], first we must be sure that
𝒃
𝑓(𝑥) is positive on it and second to find the value of the constant , 𝒌 such that ∫𝒂 𝒌. 𝒇(𝒙) 𝒅𝒙 = 𝟏 .

1
Cumulative Distribution Function 𝑭(𝒙)
The cumulative function 𝑭(𝒙) for a continuous random variable X with probability density function 𝒇(𝒙) is
𝒙
given by: 𝑭(𝒙) = 𝑷(𝑿 ≤ 𝒙) = ∫−∞ 𝒇(𝒖) 𝒅𝒖 −∞≤𝒙≤∞

Using 𝑭(𝒙) to Compute Probabilities


Let X be a CRV with pdf, 𝒇(𝒙) and CDF, 𝑭(𝒙). Then for any number a, 𝑷(𝑿 > 𝑎) = 𝟏 − 𝑭(𝒂)
and for any two numbers a and b with 𝑎 < 𝑏 , 𝑷(𝒂 ≤ 𝑿 ≤ 𝒃) = 𝑭(𝒃) − 𝑭(𝒂)

Obtaining 𝒇(𝒙) from 𝑭(𝒙)


If 𝑿 is a continuous random variable with pdf 𝒇(𝒙) and cdf, 𝑭(𝒙), then at every x at which the derivative of
𝑭(𝒙) exists, then,
𝒅𝑭
𝑭′ (𝒙) = = 𝒇(𝒙)
𝒅𝒙

Mean (or Expected Value) of a Random Variable


The Expected Value of a random variable 𝑿, denoted by 𝑬[𝑿] or 𝝁, is the average value of the random variable
𝑿 . The mean or expected value of rv 𝑿 that has a probability distribution 𝒇(𝒙) is given by:

𝝁 = 𝑬(𝒙) = ∫ 𝒙. 𝒇(𝒙) 𝒅𝒙
−∞

Variance of a Random Variable


The variance of rv 𝑿 that has pdf 𝒇(𝒙) and 𝒎𝒆𝒂𝒏 𝝁 is denoted by 𝑽𝒂𝒓(𝑿) or 𝝈𝟐 , where,

𝑽𝒂𝒓(𝑿) = 𝝈𝟐 = 𝑬(𝑿𝟐 ) − 𝝁𝟐 = ∫−∞ 𝒙𝟐 . 𝒇(𝒙) 𝒅𝒙 − 𝝁𝟐

Standard Deviation
The standard deviation 𝑺𝑫 of rv 𝑿 is the positive square root of the variance, 𝝈𝟐

𝑺𝑫 = √𝑽𝒂𝒓(𝑿) = √𝝈𝟐

Summary:

1. The mean or expected value of a random variable 𝑿 is, 𝝁 = 𝑬(𝒙) = ∫−∞ 𝒙. 𝒇(𝒙) 𝒅𝒙

2. The variance of a random variable 𝑿 is, 𝝈𝟐 = 𝑬(𝑿𝟐 ) − 𝝁𝟐 = ∫−∞ 𝒙𝟐 . 𝒇(𝒙) 𝒅𝒙 − 𝝁𝟐

3. The standard deviation of a random variable 𝑿 is, 𝝈 = √𝝈𝟐


𝒙
4. The cumulative distribution for rv 𝑿 is, 𝑭(𝒙) = 𝑷(𝑿 ≤ 𝒙) = ∫−∞ 𝒇(𝒖)𝒅𝒖 −∞≤𝒙≤∞
𝒃
5. 𝑷(𝒂 ≤ 𝑿 ≤ 𝒃) = ∫𝒂 𝒇(𝒙) 𝒅𝒙 or 𝑷(𝒂 ≤ 𝑿 ≤ 𝒃) = 𝑭(𝒃) − 𝑭(𝒂)

6. For any two numbers a and b with 𝑎 < 𝑏 𝑷(𝑿 > 𝒂) = 𝟏 − 𝑭(𝒂)

2
Examples on Continuous Random Variables
Example (1): Let X be continuous random variable having the probability density function:
𝑐𝑥 , 0≤ 𝑥≤1
𝑓(𝑥) = {
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
i) Find the value of the constant .
ii) Find the mean and variance for the continuous random variable 𝑋.
iii) Find the cumulative density function 𝐹(𝑋) and use it to evaluate 𝑃(0.2 < 𝑋 < 0.5).!!
Solution
1
∞ ∞ 1 𝑥2 1
i) Since, ∫−∞ 𝑓(𝑥)𝑑𝑥 = 1, therefore, ∫−∞ 𝑓(𝑥) 𝑑𝑥 = ∫0 𝑐𝑥 𝑑𝑥 = 𝑐 | = 𝑐 (2) = 1, ⟹ 𝑐=2
2 0

∞ 1 2 1 2
ii) The mean 𝜇 = 𝐸(𝑋) = ∫−∞ 𝑥. 𝑓(𝑥) 𝑑𝑥 = ∫0 2𝑥 2 𝑑𝑥 = 𝑥3| =
3 0 3

∞ 1 2 1 1
and 𝐸(𝑋 2 ) = ∫−∞ 𝑥 2 . 𝑓(𝑥) 𝑑𝑥 = ∫0 2𝑥 3 𝑑𝑥 = 𝑥4| =
4 0 2

1 2 2
therefore, the variance is : 𝜎 2 = 𝐸(𝑋 2 ) − 𝜇 2 = − ( 3) = 0.0556
2

iii) The cumulative distribution for the continuous random variable X with the probability
density function 𝑓(𝑥) is given by :
𝑥
𝐹(𝑥) = 𝑃(𝑋 ≤ 𝑥) = ∫−∞ 𝑓(𝑢)𝑑𝑢 −∞ ≤ 𝑥 ≤ ∞
𝑥
= ∫0 2𝑢 𝑑𝑢 = 𝑢2 |0𝑥 = 𝑥 2 0≤𝑥≤1

0 𝑥<0
Therefore, 𝐹(𝑥) = {𝑥 2 0≤𝑥≤1
1 𝑥>1

𝑃(0.2 < 𝑋 < 0.5) = 𝐹(0.5) − 𝐹(0.2) = (0.5)2 − (0.2)2 = 0.21

Example (2): Let 𝑓(𝑥), be a probability density function, where c is constant,


𝑘 √𝑥 𝑖𝑓 0 ≤ 𝑥 ≤ 1
𝑓(𝑥) = {
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Calculate, i) The value of 𝑘 ii) The mean
iii) The variance and standard deviation iv) The Cumulative distribution F(x).
v) 𝑃(0.2 < 𝑋 < 0.5) vi) 𝐹(0.2).
Solution

i) Since 𝑓(𝑥) is probability density function, then it satisfies the condition: ∫−∞ 𝒇(𝒙)𝒅𝒙 = 𝟏
∞ 1 1 𝑘 3⁄ 1 2𝑘
∵ ∫−∞ 𝑓(𝑥)𝑑𝑥 = ∫0 𝑓(𝑥) 𝑑𝑥 = ∫0 𝑘 √𝑥 𝑑𝑥 = (3 )𝑥 2|
0
= ( 3 ) = 1 , therefore, ∴ 𝑘 = 3 ⁄ 2 = 1.5
⁄ 2

3
∞ 1 1 1.5 1 1.5 3
ii) The mean 𝝁 = ∫−∞ 𝑥. 𝑓(𝑥)𝑑𝑥 = (1.5) ∫0 𝑥(𝑥 1.5 )𝑑𝑥 = (1.5) ∫0 (𝑥 1.5 )𝑑𝑥 = (2.5)𝑥 2.5 | = 2.5 (1) = 5 = 0.6
0
∞ 1 1 1.5 3
iii) ∵ 𝐸(𝑋 2 ) = ∫−∞ 𝑥 2 . 𝑓(𝑥) 𝑑𝑥 = ∫0 (𝑥 2 )(1.5𝑥 0.5 )𝑑𝑥 = 1.5 ∫0 3𝑥 2.5 𝑑𝑥 = 3.5
𝑥 3.5 |10 = 7
= 0.43

∴ The variance 𝝈𝟐 = 𝑽(𝑿) = 𝑬(𝑿𝟐 ) − 𝝁𝟐 = 0.43 − ( 0.6)2 = 0.43 − 0.36 = 0.07

And the standard deviation √𝜎 2 = √0.07 = 0.26


iv) ∵ The Cumulative distribution function 𝑭(𝒙) for the probability distribution of rv X is given by the formula:
𝒙
𝑭(𝒙) = 𝑷(𝑿 ≤ 𝒙) = ∫−∞ 𝒇(𝒖)𝒅𝒖 −∞≤𝒙≤∞
𝑥
∴ 𝐹(𝑥) = ∫0 (1.5 𝑢0.5 ) 𝑑𝑢 = 𝑢1.5 |𝑥0 = 𝑥 1.5

0 𝑥<0
𝑭(𝒙) = { 𝑥 1,5 0≤𝑥≤1
1 𝑥>1

v) 𝑃(0.2 < 𝑋 < 0.5) = 𝐹(0.5) − 𝐹(0.2) = (0.5)1.5 − (0.2)1.5 = 0.26


vi) ∵ 𝐹(𝑥) = 𝑥 1.5 ⟹ ∴ 𝐹(0.2) = (0.2)1.5 = 0.09

Example (3): The probability density function of a random variable X, that represents the lifetime of certain
type of electronic devices (measured in hours), is given by,
10/𝑥 2 𝑖𝑓 𝑥 > 10
𝑓(𝑥) = {
0 𝑥 ≤ 10

i) What is the cumulative function of the probability distribution of 𝑋?


ii) Find 𝑃(20 < 𝑋 < 40) and 𝑃(𝑋 > 20).
iii) What is the probability that of 6 such types of devices at least 3 will function for at least 15 hours? What
assumptions are you making?
Solution
i) The cumulative function of the probability distribution of 𝑋 is given by :
𝑥
𝐹(𝑥) = 𝑃(𝑋 ≤ 𝑥) = ∫−∞ 𝑓(𝑢)𝑑𝑢 −∞≤𝑥 ≤∞
𝑥 10 1 𝑥 1 1 10
∴ 𝐹(𝑥) = ∫10 ( 𝑢2 ) 𝑑𝑢 = − 10 |𝑢 | = −10 [𝑥 − 10] = [1 − ]
10 𝑥

𝟏𝟎
𝟏− 𝒊𝒇 𝒙 > 𝟏𝟎
𝒙
Therefore, the cumulative distribution is; 𝑭(𝒙) = {
𝟎 𝒙 ≤ 𝟏𝟎

𝟏𝟎 𝟏𝟎 3 1 1
ii) 𝑃(20 < 𝑋 < 40) = 𝐹(40) − 𝐹(20) = [𝟏 − 𝟒𝟎] − [𝟏 − 𝟐𝟎] = 4 − 2 = 4 = 0.25 and,

∞ ∞ ∞ 1 1
𝑃(𝑋 > 20) = ∫20 𝑓(𝑥)𝑑𝑥 = ∫20 (10/𝑥 2 ) 𝑑𝑥 = 10 ∫20 (𝑥 −2 ) 𝑑𝑥 = −10| 𝑥−1 |∞
20 = −10 [0 − 20] = 2

4
iii) What is the probability that of 6 such types of devices at least 3 will function for at least 15 hours? What
assumptions are you making? This is another one of those problems that must be considered hierarchically in
steps. Break the problem up into two steps:
a) finding the probability that 1 device functions for at least 15 hours, and
b) finding the probability that 3 of the 6 devices function for at least 15 hours.
Step 1: One Device
First we must find the probability that one randomly selected device will function for at least 15 hours.
It is given that each device follows a probability distribution 𝒇(𝒙) given earlier. Let 𝑋 be the number of hours
that the device functions. Then,
∞ ∞ ∞ 1 2
𝑃(𝑋 ≥ 15) = ∫15 𝑓(𝑥)𝑑𝑥 = ∫15 (10/𝑥 2 ) 𝑑𝑥 = 10 ∫15 (𝑥 −2 ) 𝑑𝑥 = −10| 𝑥−1 |∞
15 = −10 [0 − 15] = 3

Step 2: 3 of the 6 Devices


We are interested in 𝒏 = 𝟔 devices. Note that each device constitutes a trial with one of two outcomes: device
functions at least 15 hours or it doesn’t. Each of the 6 trials are Bernoulli trials and we assume the trials are
independent. We also have a constant probability of success 𝒑 = 𝟐/𝟑.
We use the binomial model: 𝑷(𝑿) = 𝒏𝑪𝒓 × 𝒑𝒓 × 𝒒𝒏−𝒓 𝒓 = 𝟎, 𝟏, 𝟐, 𝟑, … … . , 𝐧
Now let X be the number of devices that function at least 15 hours. Then,
𝑷(𝑿 ≥ 𝟑) = 𝑷(𝑿 = 𝟑) + 𝑷(𝑿 = 𝟒) + 𝑷(𝑿 = 𝟓) + 𝑷(𝑿 = 𝟔)
2 3 1 3 2 4 1 2 2 5 1 1 2 6 1 0
= [6𝐶3 × ( ) × ( ) ] + [6𝐶4 × ( ) × ( ) ] + [6𝐶5 × ( ) × ( ) ] + [6𝐶6 × ( ) × ( ) ]
3 3 3 3 3 3 3 3

𝟏𝟔𝟎 𝟐𝟒𝟎 𝟏𝟗𝟐 𝟔𝟒 𝟔𝟓𝟔 𝟔𝟓𝟔


= 𝟕𝟐𝟗
+ 𝟕𝟐𝟗
+ 𝟕𝟐𝟗
+ 𝟕𝟐𝟗
= 𝟕𝟐𝟗
∴ 𝑷 (𝑿 ≥ 𝟑 ) = ≈ 𝟎. 𝟗
𝟕𝟐𝟗

Example (4) The time unit a chemical reaction is complete (in millisecond) is approximated by the cumulative
distribution function,
1 − 𝑒 −0.01 𝑥 𝑥≥0
𝑭(𝒙) = {
0 𝑥<0
a) Determine the probability density function of X .
b) What proportion of the reaction is complete within 200 milliseconds ?
Solution
a) Using the result that the probability density function is the derivative of the 𝐹(𝑥), i.e

𝒅 𝟎. 𝟎𝟏𝒆−𝟎.𝟎𝟏 𝒙 𝒙≥𝟎
∵ 𝒇(𝒙) = 𝑭(𝒙) , we obtain; 𝒇(𝒙) = {
𝒅𝒙
𝟎 𝒙<𝟎
b) The probability that a reaction completes within 200 milliseconds is
𝑃(𝑋 < 200) = 𝐹(200) = 1 − 𝑒 (−0.01 )×(200) = 1 − 𝑒 −2 = 0.8647 .

5
Some Important Continuous Distributions
There are a number of continuous distributions which have important applications in engineering and science.
Among the most important continuous probability distributions are:
(1) The Uniform Distribution
This Section introduces the simplest type of continuous uniform distribution which features a continuous
random variable 𝑋 with probability density function 𝑓(𝑥) which assumes a constant value over a finite interval.
Definition: The Uniform (or Rectangular) Distribution of random variable 𝑿 restricted to a finite interval [𝑎, 𝑏]
has a constant probability density function 𝒇(𝒙) = 𝒌 defined by:
𝒌 𝒊𝒇 𝒂 ≤ 𝒙 ≤ 𝒃
𝒇(𝒙) = {
𝟎 𝒐𝒕𝒉𝒆𝒓𝒘𝒊𝒔𝒆

Example (1): A clock stops at random at any time during the day. If 𝑿 is the time (hours plus fractions of
hours) at which the clock stops.
i) Determine the probability density function for 𝑋.
ii) Calculate the expected value and variance for 𝑋 at which the clock stops.
iii) Find the cumulative distribution function for rv 𝑋.
iv) Find the probability that the clock will stop between 2:00 pm and 2:45 pm using two methods.
Solution
∵ 𝑿 is the time at which the clock stops and follows a uniform distribution over the interval [0 , 24].

Therefore, its probability density function for 𝑋 is 𝑓(𝑥) = 𝑘 and it satisfies; ∫−∞ 𝒇(𝒙)𝒅𝒙 = 𝟏


∞ 𝟐𝟒 𝟐𝟒
𝒅𝒙 = 𝒌. | 𝑥 |
24 = 𝑘(24 − 0) = 24𝑘
∫−∞ 𝒇(𝒙)𝒅𝒙 = ∫𝟎 𝒌 𝒅𝒙 = 𝒌. ∫𝟎
0
𝟏
∴ 24𝑘 = 1 ⟹ 𝒌 = 𝟐𝟒

i) Over the interval [0 , 24] the probability density function 𝑓(𝑥) is given by the formula:
𝟏
𝒇(𝒙) = { 𝟐𝟒 𝒊𝒇 𝟎 ≤ 𝒙 ≤ 𝟐𝟒
𝟎 𝒐𝒕𝒉𝒆𝒓𝒘𝒊𝒔𝒆

ii) The expected value of 𝑋 is, 𝑬(𝑿) = ∫−∞ 𝒙. 𝒇(𝒙) 𝒅𝒙
𝑏 𝟏 24 𝟏 𝑥2 24 𝟏 𝟏
Therefore, 𝑬(𝑿) = ∫𝑎 𝑥. 𝑓(𝒙) 𝑑𝑥 = ( ) ∫0 𝑥. 𝑑𝑥 = ( ) . | | = ( ) ( ) 242 = 12
𝟐𝟒 𝟐𝟒 2 0 𝟐𝟒 𝟐

The variance of 𝑋 is, 𝑽𝒂𝒓(𝑿) = 𝑬(𝑿𝟐 ) − [𝑬(𝑿)]𝟐 = ∫−∞ 𝒙𝟐 . 𝒇(𝒙) 𝒅𝒙 − [𝑬(𝑿)]𝟐
24 𝟏 1 𝑥3 24 (12)2 (24)2
∴ 𝑽𝒂𝒓(𝑿) = ∫0 𝒙𝟐 . (𝟐𝟒) 𝑑𝑥 − (12)2 = | | − = − (12)2 = 48
(24) 3 0 3

iii) The cumulative distribution function 𝑭(𝒙) for the continuous random variable 𝑿 is defined by:
𝒙
𝑭(𝒙) = 𝑷(𝑿 ≤ 𝒙) = ∫−∞ 𝒇(𝒖) 𝒅𝒖 𝟎 ≤ 𝒙 ≤ 𝟐𝟒
𝑥 𝟏 𝟏 𝟏
∴ 𝐹(𝑥) = ∫0 (𝟐𝟒) 𝑑𝑢 = (𝟐𝟒) . | 𝑢 |𝑥0 = (𝟐𝟒) . 𝑥
6
𝟎 𝒙<𝟎
𝒙
𝑭(𝒙) = { 𝟎 ≤ 𝒙 ≤ 𝟐𝟒
𝟐𝟒
𝟏 𝒙 > 𝟐𝟒
14.75 14 0.75 1
iv) 𝑷(𝟏𝟒 < 𝑿 < 𝟏𝟒. 𝟕𝟓) = 𝑭(𝒃) − 𝑭(𝒂) = 𝐹(14.75) − 𝐹(14) = ( ) − (24) = = 32 and also,
24 24
𝒃 14.75 14.75 1 1 1
𝑷(𝟏𝟒 ≤ 𝑿 ≤ 𝟏𝟒. 𝟕𝟓) = ∫𝒂 𝒇(𝒙)𝒅𝒙 = ∫14 𝑓(𝑥)𝑑𝑥 = ∫14 (24) 𝑑𝑥 = (24) [14.75 − 14] = 32

Example (2): The thickness x of a protective coating applied to a conductor designed to work in corrosive
conditions follows a uniform distribution over the interval [20,40] microns. Find the mean, standard deviation
and cumulative distribution function of the thickness of the protective coating. Find also the probability that the
coating is less than 35 microns thick.
Solution
∵ Let 𝑋 be the random variable that represent the thickness x of a protective coating applied to a conductor
and follows a uniform distribution over the interval [20 , 40]. Therefore, its 𝑓(𝑥) is pdf that satisfies;
∞ ∞ 𝟒𝟎 𝟒𝟎 40 = 𝑘(40 − 20) = 20𝑘
∫−∞ 𝒇(𝒙)𝒅𝒙 = 𝟏, ⟹ ∫−∞ 𝒇(𝒙)𝒅𝒙 = ∫𝟐𝟎 𝒌 𝒅𝒙 = 𝒌. ∫𝟐𝟎 𝒅𝒙 = 𝒌. | 𝑥 |
20
𝟏 𝟓
∴ 20𝑘 = 1 ⟹ 𝒌 = 𝟐𝟎 = 𝟏𝟎𝟎 = 𝟎. 𝟎𝟓

i) Over the interval [20 , 40] the probability density function 𝑓(𝑥) is given by the formula:
𝟎. 𝟎𝟓 𝒊𝒇 𝟐𝟎 ≤ 𝒙 ≤ 𝟒𝟎
𝒇(𝒙) = {
𝟎 𝒐𝒕𝒉𝒆𝒓𝒘𝒊𝒔𝒆
∞ 40 𝑥2 40 (𝟎.𝟎𝟓) [402
ii) Mean 𝑬(𝑿) = ∫−∞ 𝒙. 𝒇(𝒙) 𝒅𝒙 = (𝟎. 𝟎𝟓) ∫20 𝑥. 𝑑𝑥 = (𝟎. 𝟎𝟓). | | = 2 − 202 ] = 30 𝜇𝑚
2 20
∞ 40
The variance 𝑽𝒂𝒓(𝑿) = 𝑬(𝑿𝟐 ) − [𝑬(𝑿)]𝟐 = ∫−∞ 𝒙𝟐 . 𝒇(𝒙) 𝒅𝒙 − [𝑬(𝑿)]𝟐 = ∫20 𝑥2 . (0.05) 𝑑𝑥 − (30)2
𝑥3 40 (30)2 (𝟎.𝟎𝟓)×[403 − 203 ]
= (0.05) | | − = − (30)2 ≈ 33.33
3 20 3

The standard deviation √𝝈𝟐 ≈ √33.33 ≈ 5.77 𝜇𝑚


The cumulative distribution function 𝑭(𝒙) for the continuous random variable 𝑿 is defined by:
𝒙
𝑭(𝒙) = 𝑷(𝑿 ≤ 𝒙) = ∫−∞ 𝒇(𝒖)𝒅𝒖 𝟐𝟎 ≤ 𝒙 ≤ 𝟒𝟎
𝑥 𝑥
∴ 𝑭(𝒙) = ∫20(0.05) 𝑑𝑢 = (0.05). | 𝑢 |20 = (0.05)[𝑥 − 20]
𝟎 𝒙 < 𝟐𝟎
𝒙
𝑭(𝒙) = { −𝟏 𝟐𝟎 ≤ 𝒙 ≤ 𝟒𝟎
𝟐𝟎
𝟏 𝟒𝟎 < 𝒙
𝑷(𝑿 < 𝟑𝟓) = 𝐹(𝑿 ≤ 𝟑𝟓) = (0.05)[35 − 20] = (0.05) × 15 = 𝟎. 𝟕𝟓 and in other method,
𝟑𝟓 35 35
𝑷(𝑿 < 𝟑𝟓) = ∫𝟐𝟎 𝒇(𝒙) 𝒅𝒙 = (0.05) ∫20 𝑑𝑥 = (0.05). | 𝑥 | = (0.05)[35 − 20] = (0.05) × 15 = 𝟎. 𝟕𝟓
20
7
Problem: The current (in mA) measured in a piece of copper wire is known to follow a uniform distribution over
the interval [0,25]. Write down the formula for the probability density function 𝑓(𝑥) of the random variable X
representing the current. Calculate the mean, variance of the distribution, the cumulative distribution function
𝐹(𝑥) and then calculate 𝑃(15 < 𝑋 < 20) using two methods.

(2) The Exponential distribution


Introduction: The exponential distribution is one of the widely used continuous distributions. It is often used to model
the time elapsed between events. If an engineer is responsible for the quality of, say, copper wire for use in
domestic wiring systems, he or she might be interested in knowing both the number of faults in a given length
of wire and also the distances between such faults. While the number of faults may be analyzed by using the
Poisson distribution, the distances between faults along the wire may be shown to give rise to the exponential
distribution defined and used in this Section.

Definition: A continuous random variable 𝑋 is said to have an exponential distribution with parameter 𝜆, if its
pdf is given by
𝝀𝒆−𝝀𝒙 𝑥 ≥ 0,
𝒇(𝒙) = {
𝟎 𝒐𝒕𝒉𝒆𝒓𝒘𝒊𝒔𝒆
where 𝝀 > 𝟎 is called the rate of the distribution.

Example (1): 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.
Solution
Let 𝑿 be the random variable “time to failure” (in hours), and since the random variable X has an exponential
distribution, therefore its pdf could be written in the form: 𝒇(𝒙) = 𝝀𝒆−𝝀𝒙 . Where, 𝝀 is parameter.
(a) First we need to find the value of 𝝀
1000
∵ 𝑃(𝑋 ≤ 1000) = ∫0 𝜆 𝑒 −𝜆𝑥 𝑑𝑥 (1) and 𝑃(𝑋 ≤ 1000) = 10 % = 0.1 (2)

(−𝜆)𝑒 −𝜆𝑥 𝑑𝑥 = −| 𝑒 −𝜆𝑥 |1000 = −[ 𝑒 −1000𝜆 − 1] = 1 − 𝑒 −1000𝜆


1000 1000
∫0 𝜆 𝑒 −𝜆𝑥 𝑑𝑥 = − ∫0
0
From (1) and (2) we get;
1 − 𝑒 −1000𝜆 = 0.1 ⟹ 0.9 = 𝑒 −1000𝜆 ⟹ 𝑙𝑛(0.9) = 𝑙𝑛( 𝑒 −1000𝜆 )
0.10536
∴ −0.10536 = −1000 𝜆 ⟹ 𝜆= ⟹ 𝜆 ≈ 1.05 × 10−4
1000

If 𝑿 is the time till failure, the question asks for 𝑃(𝑋 > 5000):
∞ ∞ ∞
𝑃(𝑋 > 5000) = ∫𝟓𝟎𝟎𝟎 𝝀 𝒆−𝝀𝒙 𝒅𝒙 = − ∫𝟓𝟎𝟎𝟎(−𝝀) 𝒆−𝝀𝒙 𝒅𝒙 = −| 𝒆−𝝀𝒙 | = −[0 − 𝒆−𝟓𝟎𝟎𝟎𝝀 ] ≈ 0.59
5000

8
(b) Find the mean and standard deviation of the time till failure.
∞ ∞
𝑴𝒆𝒂𝒏, is, 𝝁 = ∫−∞ 𝒙. 𝒇(𝒙) 𝒅𝒙 = ∫0 𝑥. (𝝀𝒆−𝝀𝒙 ) 𝑑𝑥 Using Integration by parts,

Let 𝑢 = 𝑥 𝑑𝑣 = 𝝀 𝒆−𝝀𝒙
𝒅𝑢 = 𝑑𝑥 𝑣 = − 𝒆−𝝀𝒙
∞ ∞ ∞ ∞ 1 ∞ 1 𝟏
∴ 𝝁 = ∫0 𝑥. (𝝀𝒆−𝝀𝒙 ) 𝑑𝑥 = − | 𝑥 𝒆−𝝀𝒙 | + ∫0 (𝒆−𝝀𝒙 ) 𝑑𝑥 = − | 𝑥 𝒆−𝝀𝒙 | − . | 𝒆−𝝀𝒙 | = − [0 − 1] =
0 0 𝝀 0 𝝀 𝝀
𝟏 𝟏𝟎𝟎𝟎
𝑴𝒆𝒂𝒏 𝝁 = = = 𝟗𝟒𝟗𝟏. 𝟐𝟕 𝒉𝒐𝒖𝒓𝒔
𝝀 𝟎.𝟏𝟎𝟓𝟑𝟔

As an exercise, you may wish to verify that by applying integration by parts twice,

𝑬(𝑿𝟐 ) = ∫−∞ 𝒙𝟐 . 𝒇(𝒙) 𝒅𝒙 = ⋯ = 𝟐/𝝀𝟐
𝟐 𝟏 𝟐
The variance of 𝑋 is, 𝑽𝒂𝒓(𝑿) = 𝑬(𝑿𝟐 ) − [𝑬(𝑿)]𝟐 = 𝟐 − 𝟐 = 𝟏/𝝀
𝝀 𝝀
24 𝟏 1 𝑥3 24 (12)2 (24)2
∴ 𝑽𝒂𝒓(𝑿) = ∫0 𝒙𝟐 . (𝟐𝟒) 𝑑𝑥 − (12)2 = | | − = − (12)2 = 48
(24) 3 0 3

𝟏 𝟏
𝑺𝒕𝒂𝒏𝒅𝒂𝒓𝒅 𝒅𝒆𝒗𝒊𝒂𝒕𝒊𝒐𝒏 = √𝑽𝒂𝒓𝒊𝒂𝒏𝒄𝒆 = √𝝀𝟐 = 𝝀 = 𝟗𝟒𝟗𝟏. 𝟐𝟕 𝒉𝒐𝒖𝒓𝒔

Example (2): Reliability


Traditional electric light bulbs are known to have a mean lifetime to failure of 2000 hours. It is also known that
the distribution function 𝑓(𝑡) of the time to failure takes the form 𝑓(𝑡) = 1 − 𝑒 −𝑡/µ , where µ is the mean time
to failure. You will see if you study the topic of reliability in more detail that this is a realistic distribution
function. The reliability function 𝑟(𝑡), giving the probability that the light bulb is still working at time t, is
defined as 𝑟(𝑡) = 1 − 𝑝(𝑡) = 𝑒 −𝑡/µ Find the proportion of light bulbs that you would expect to fail before
1500 hours and the proportion you would expect to last longer than 2500 hours.
Solution
Let 𝑻 be the random variable ‘time to failure’.
The proportion of bulbs expected to fail before 1500 hours is given as:
𝑃(𝑋 < 1500) = 1 − 𝑒 −1500/2000 = 1 − 𝑒 −3/4 = 1 − 0.4724 = 0.5276
The proportion of bulbs expected to last longer that 2500 hours is given as:
𝑃(𝑇 > 2500) = 1 − 𝑃(𝑇 < 2500) = 𝑒 −2500/2000 = 𝑒 −5/4 = 0.2865
Using 𝒓(𝒕) = 𝟏 − 𝒑(𝒕) we have, 𝑟(2500) = 0.2865.
Hence we expect just under 53% of light bulbs to fail before 1500 hours service and just under 29% of light bulbs
to give over 2500 hours service.

Dr.Mohamed Said

9
MSA University Module Title: Probability and Statistics
Faculty of Engineering Module Code: MAT 361/231

Exercise on Continuous Probability Distributions


(1) Check whether of the following functions are suitable probability density functions over the given range:

1 1
a) 𝑓(𝑥) = 2 (𝑥 2 + 4) 0 ≤ 𝑥 ≤ 1, b) 𝑓(𝑥) = 12 (2𝑥 + 1) 0 ≤ 𝑥 ≤ 3,
10
c) 𝑓(𝑥) = 𝑥 2 − 4𝑥 + 0 ≤ 𝑥 ≤ 3, d) 𝑓(𝑥) = 6𝑥(1 − 𝑥) 0≤𝑥≤1
3
𝑥
e) 𝑓(𝑥) = 4 1≤𝑥≤3 f) 𝑓(𝑥) = 12𝑥(1 − 𝑥)2 0≤𝑥≤1

(2) Normalize each of the following functions over the given range:
𝑒
a) 𝑓(𝑥) = 𝑘𝑒 −𝑥 0≤𝑥≤1 Ans: 𝑘 = ≈ 1.582
𝑒−1

b) 𝑓(𝑥) = 𝑘𝑥(6 − 𝑥 2 ) 0<𝑥<6 Ans: 𝑘 =?

(3) Let 𝑓(𝑥), be a probability density function, where,


𝑘𝑥 2 𝑖𝑓 0 ≤ 𝑥 ≤ 1
𝑓(𝑥) = {
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Calculate:
i) The value of the constant 𝒌, ii) The mean, variance and standard deviation.
iii) The cumulative distribution F(x). iv) 𝑃(𝑋 < − 1.5), 𝑃(𝑋 > 1.5) and 𝑃(0.2 < 𝑋 < 0.5).

(4) Let 𝑓(𝑥), be a probability density function, where k is constant, where


𝑘𝑥(2 − 𝑥) 0<𝑥<2
𝑓(𝑥) = {
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒.
Calculate:
i) The value of k, ii) The mean iii) The variance and standard deviation
2
iv) The Cumulative distribution F(x). v) 𝑃(1 < 𝑋 < 1.5) vi) 𝑃( 𝑋 < 3 )

(5) The time unit a chemical reaction is complete (in millisecond ) is approximated by the cumulative distribution
function,
0 𝑥<0
𝐹(𝑥) = {
1 − 𝑒 −0.01 𝑥 0≤𝑥
i) Determine the probability density function of X.
ii) What proportion of the reaction is complete within 200 milliseconds?

(6) The lifetime T (years) of an electronic component is a continuous random variable with a probability density function
given by; 𝒇(𝒙) = 𝒆−𝒙 𝑥 ≥ 0,
Find the lifetime L which a typical component is 60% certain to exceed. If five components are sold to a manufacturer
find the probability that at least one of them will have a lifetime less than L years.

10

You might also like