DEPARTMENT OF ECE

TECHNICAL REPORT

APPLICATION OF CONTINUOUS
PROBABILITY DISTRIBUTON

SWETA GHOSH
2ND YEAR,3RD SEM;
10900323119

NETAJI SUBHAS ENGINEERING COLLEGE

 Application of
Continuous Probability
Distribution
Structure
1. Introduction

Objectives

2. Continuous Random
Variables
Probability Distributions
Mean & Variance

3. Types of Continuous Distributions

Uniform Continuous Distribution
Exponential Distribution

4. Normal Distribution

1. INTRODUCTION
Let us turn our attention to a variable which takes all the values within a given range or an interval. Such a variable is
called a continuous random variable. Measurements of height, temperature, amount of rainfall and waiting time are some
of the examples of such a variable. Consider, for instance, the following situation: A professor of statistics catches a bus
every evening to take him home from his university. He has never bothered to find out the exact timings of the departure
of the buses run since there is a frequent and punctual service- a bus leaves after every ten minutes. When his work for
the day in the University is over, he walks down to the bus stand and catches the next bus. One hot evening as he is
waiting for the bus he wonders if he would save much time and discomfort if he could find out the bus timings so that he
could time his arrival at the bus stand in a better way. He decides to investigate his problem by setting up a mathematical
model. If he continues his usual practice, he is equally likely to wait any time from 0 minute to 10 minutes. He may
arrange his waiting time by a discrete random variable which can take the ten, values say 0.5, 1.5, 2.5 ,..., 9.5. If he wants,
he can record the timings to the nearest minute or half minute. So he may be justified to use a discrete random variable.
But the professor thinks he will do better (or will have a better model) if the waiting-time variable could take any value
within a ten minute interval i.e. the interval or range 0-10. This waiting time variable is not a discrete random variable
since it can take any value from 0 to 10 say 8.7754. Therefore, the professor decides to choose a continuous random
variable which can take continuously values. Such a random variable, as you know from Unit 11, is called a continuous
random variable. It, therefore, follows that there are situations where the discrete random variable does not help and we
have to bring in the concept of a continuous random variable. What is, then, the corresponding probability distribution of
a continuous random variable? How to calculate the mean, variance and other measures of a continuous random
variable? We shall try to find answers for such questions in this unit.
 Objectives:
• describe a probability distribution of a continuous random variable;
• calculate, the mean and variance of a continuous random variable;
• discuss some special types. of continuous distributions.

2. Continuous Random Variables

Continuous random variables represent measured data such as all possible heights, weight's, temperatures, rainfalls,
distances, life periods, whereas, discrete random variables represent count data such as number of children in a family,
number of defective bulbs in an electrical firm, Dr number of accidental deaths in a year and so on. In short, a
continuous random variable.i~ one that takes all values within a given range on a continuous scale. The outcomes of an
experiment in a continuous case are represented by the points on a line. Anyone can ask, "Are there only two random
variables, discrete and continuous"? No. Net all, random variables are either discrete or continuous. In fact, there are
random variables which are neither discrete nor continuous. Also, there are random variables which are partly discrete
and partly continuous.

PROBABILIT2.Y DISTRIBUTIONS
Let us now discuss the question of assigning or distributing probability values to the values of a continuous random
variable. .While doing so, you may recall that i) a probability value p assigned to a value of a random variable always lies
between 0 and 1 i.e., 0<P<1; ii) the sum of all such probability values is equal to 1. A probability value p assigned to a
value of a random variable always lies between 0 and 1 i.e.. 0 < 1 the sum of all such probability values is equal to 1 (see
unit 12). In Unit 13, you have learnt the methods of assigning or distributing. Probability values to the values of a
random variable. Take, now, the case of a continuous random variable. For example, consider a random variable whose
values are the heights of all persons over 20 years of age. Between any two values; say 168.5 cm and 169.5 cm, there
are infinite number of heights, of which only one is 169 centimeters. The probability of selecting a person at random
who is exactly 169 cm tall is extremely remote and hence will be zero. In other words, we assign a probability of zero to
that event. Thus la continuous random variable has a probability value zero of assuming exactly any of its values. This,
however, is not the case if we talk about the probability of selecting a person who is atleast 168 cm but not more than
170 cm tall with an interval rather than a point value of a random variable. This example tells us that we cannot define a
probability distribution in the same way as for the discrete random variable where we assign non-zero probabilities for a
random variable X taking particular values. But we can see that in the case of a continuous random bariable X, the
probability that X lies between two values is meaningful. That is, there has to be a different method of assigning
(distributing) probabilities to the values of a continuous random variable for various intervals such as
P(a<X<b),P(X>c),P(X<d) etc.
where X is a continuous random variable taking values in an interval (a, b) or X takes all values greater than a number c
or takes all values less than a number d. It does- not matter whether we include in the interval the end points a and b or
not, that is, to say, the interval may be open or closed or semi-open or semi- closed. This is because of the following
reason: Since a conlinuous random variable has a probability of zero of assuming an exact value, therefore \ P (X = a) =
0, P (X = b) = 0. Hence,
P(aCX≤b)=P(X=a)+P(a<X<b)+P(X=b)=0+P(a<X<b)+1=P(a<X<b)
thus when X is a continuous random variable, then we have
P(a≤X≤b)=P(a<X≤b)=P(a≤X<b)= P(a<X<b)
Similarly, P(X ≥c)=P(X>c) (X ≤d )=P(X<d)
Thus in the case of a continuous random variable X, it makes no difference if X takes values in a closed,interval or in an
open interval or in a semi-closed or semi- open interval. In view of this, we shall use the interval (a, b) for all types of
intervals.
The probabilities associated with values of a continuous random variable will be represented by the areas bounded
by the lines/curves like the ones give? Also, these probabilities are positive numerical values. Therefore, the graph of
the density function f(x) must lie above the X-axis and between an interval (a, b). Moreover, a probability distribution
is such that the sum of the probabilities in the distribution is 1. Therefore, the total area under its curve bounded by
the X-axis is equal to 1.

I. Probability Density Function (PDF) :

A function with values f(x) is called a probability density function (PDF) for the continuous random variable X if
i) f(X) ≥0
ii) the total area under its curve and above the X-axis is equal to 1,
iii) the area under the curve between any two ordinates ,x = a and x = b and the X-axis gives the probability that X lies
between a and b i.e. P (a 5 x s b) = area under the curve between x = a and x = b and the X-axis.
Consider the following example of a density function:
Example 1: A continuous random variable X that can assume values between x= 2 and x = 4 has a density function f(x)
given by. F(x)=(x+1)/8; find P(2<X<3).

Solution: As soon as you draw the graph of the given function, you will find that the region under the given
restrictions namely 2 < X < 3 is the region as shaded in the Figure.

This shaded region represents P (2 < X < 3) and hence P(2 < X < 3) = Area of the shaded region. The shaded region is a
trapezium whose area is found by summing up the parallel heights, multiplying the sum by the length of the base and
dividing by 2 i.e.
P(2<X<4)=(sum of parallel sides) X base = [f(2)+f(3)]X 2 =3/8+4/8 X 2 =7/8.
2 2 2
II. The Cumulative Distribution Function (CDF) Let X be a continuous random variable with PDF f(x). Then the
𝜘
probability that X has a value less than or equal to x is given by F(x)=P(X ≤x)=‫׬‬−∞ 𝑓 𝑥 ⅆ𝑥. F(x) is called the cumulative
dF(x)/dx=f(x) or, F’(x)=f(x)
∴ F(-∞)=0 , F(∞)=1
and F(x) is a non-decreasing function of x, that is,if y ≥ x, then, F(y) ≥F(x); also F’(x)=f()x≥0

MEAN & VARIANCE:

The Mean 𝜇𝜘 or the expected value E(Xof a Continuous Random Variable with PDF f(x) is defined as follows:
∞ ∞ 𝑏
x= 𝜇𝜘 =E(x)=‫׬‬−∞ 𝜘 ⅆ𝑥 for -∞<x< ∞ while ‫׬‬−∞ 𝜘 ⅆ𝑥=1 or, x= 𝜇𝜘 =E(x)=‫ 𝜘 𝑥 𝑎׬‬ⅆ𝜘 for a<x<b
𝑥

• The Variance of a Continuous Random Variable The variance V(x) of a continuous random variable X with PDF f(x) is
defined as follows:
∞ ∞
V(X)= ‫׬‬−∞(x− 𝜇𝜘 )^2 𝜘 ⅆ𝑥= ‫׬‬−∞ x^2 𝜘 ⅆ𝑥- 𝜇𝜘 ^2
The variance is a measure of the dispersion (or scatter.) of the values of X. Its square root 𝑣 𝑥 is called the standard
deviation of X and is denoted by 𝜎𝜘.

3. TYPES OF CONTINUOUS DISTRIBUTION:

I. Uniform Continuous Distribution :
A uniform distribution is one in which the probability values (frequencies) are equally or uniformly distributed.
Continuous random variable is said to be uniformly distributed in the interval (a, b) if it is equallylikely to lie anywhere
in this interval but cannot lie outside it. In other words, it is the probability distribution in which type CRV assumes all
its values in an interval with equal probabilities. For example, if we measure the height in inches of a group of men
and consider only the fractional part of each height ignoring the integral part, then the result will be a randoni
variable which must take all values between 0 and 1. The probability of this variable will be nearly evenly or equally
distributed over this interval. Moreover, in this example the density function of meha random variable is always a
constant which must be 1 since the area under the density function must be 1. Hence, f(x)=1, 0≤ x ≤ 1
= 0, otherwise.
is the density function in this case. The integration of this function with respect to x gives us the cumulative
distribution function F(x) as- F(x)=x, 0≤ x ≤ 1
=0, x ≤ 0
=1, x≥1
In general, the density function of a continuous random variable X with uniform probability distribution is given by
f(x)=1/(b-a); a≤ x ≤ b, a≠b
=0; otherwise

The CDF of X is given by

II. Exponential Distribution :
A number of real life situations describe another kind of continuous probability distribution. This is called an
exponential distribution. Suppose we have a collection of electric bulbs made in the same factory with the same
specification. We may record the length of the time each bulb lasts before it fails. Some of the bulbs will last a short
life and some will last a long time. This provides us a model of an exponential distribution. Consider another quite
familiar situation. Suppose an infectious disease like cholera has spread in a village. What is the length of time
between successive reports of the disease when it is spreading in a random manner through the. population? This is
described by a continuous variable that can be modelled using the exponential distribution. The word exponential is
derived from the exponential function as is evident from the probability density function of the exponential
distribution.
If f(x)=∞e^(−𝑎𝑥), ∞<0, x≥0
= 0, otherwise
where ∞ is some real number 'called the parameter of the distribution. We Say that a random variable X with PDF
f(x) is exponentially distributed with parameter a and hence is sometimes called an exponential variable. The range of
the variable is (0, ∞ ). The graph of f(x) is shown below in Figure.

1. NORMAL DISTRIBUTION:
Suppose a weather expert is to measure the temperature of a day during a particular interval of time say 10 A.M. to
2.00 P.M. Assume that his highest reading of 12 noon is 41°C. All the readings when graphed on a paper from a curve
like the one given in Figure

Such a curve is a bell-shaped curve and is symmetrical about the line at the centre (see Fig. ). This is called a normal
curve and the distribution of the scores (which gave the above curve) is called a Normal Distribution.

The word 'normal' is one of the most fascinating words in statistics. We often come across the word 'normal' in our
daily life. The doctor tells that the condition of the patient is 'normal'. The situation in the city after some violence is
back to normal. We have other usages of the word 'normal' viz normal hydrocarbons in chemistry, normal ranges in
medicine, the water level in the river is above 'normal' due to floods, the humidity is below 'normal' etc. Thus the word
conveys two senses namely something desirable or something commonly found.
We may, however, say that the word 'normal, means the desirability of what is commonly found so that the two senses
reinforce each other.
The normal distribution is the most important continuous probability distribution in the entire area of statistics. It is one
of the three main theoretical distributions. The other two important theoretical distributions as we know , are binomial
and Poisson. The binominal distribution was proposed in 1700 by Jacob Bernoulli (1654- 1705), a Swiss mathematician
and the Poisson distribution was given in 1837 by S.D. Poisson (1781 - 1840), a French mathematician. The normal
distribution is due to De Moivre (1667 - 1754). another French mathematician who is also known for his popular De
Moivre's theorem in trigonometry. However, the normal distribution is more commonly associated with the later
mathematicians Gauss (1777 - 1855), a German and Laplace (1749- 1827), a French. That is why physicists or engineers
often call it the Gaussian distribution but in France, it is called Laplacean distribution. K. Pearson (1857 - 1936) a British
mathematician, appears to have coined the name 'normal'. The normal distribution is, without doubt, the most important
distribution in theoretical statistics.
Definition: A continuous random variable X is said to have a normal distribution if its probability density function is of the
form
𝑥−𝜇 2
f(x)=1/ 2𝜋𝜎 e − σ >0, -∞<x< ∞
2𝜎2

The density function here consists of the two parameters p (mean) and a (standard deviation). The variable X is generally
called a. normal random variable. The density function f(x) is also denoted by N (𝜇, 𝜎2).

The graph of the density function f(x) is a bell shaped curve or like a cocked-hat as shown in the Figure

The curve is generally called a normal curve. The normal curve has the following important properties:
• The mean and variance of a normal variable are 𝜇 and 𝜎. Once the mean 𝜇 and the standard deviation 𝜎 are
specified, then the normal curve is completely determined. For example, if 𝜇 = 50, 𝜎 = 5 then the ordinates of f(x)
can be easily computed for various values of x and the curve can be drawn.
• The curve attains its maximum at the point x = 𝜇 . In other words, the function f(x) is maximum for x = 𝜇 , i.e. at
the mean.
• The curve is symmetrical about a vertical axis through the point x = 𝜇 i.e. through the mean.For example, if x =
p+3, then also 𝜘 − 𝜇 2 = 9.In both the cases f(x) remains the same.
• The curve approaches infinity along the horizontal axis in either direction from the mean.
• The total area under the curve and above the horizontal axis is equal to 1. In view of this property, the area
under the curve between two ordinates say X = a and X = b, a < b represents.The probability that X lies between a
and b, denoted by P(a < X < b) as shown in the Figure.

To find the probability distribution of a normal random variable, we have to calculate the corresponding areas
carved by the normal curves. The shapes and hence the areas under the normal curves will be different according to
the different values of the mean p and the standard deviation a. In this way, it becomes a tedious task to sketch
separate curves for every possible values of p and a and hence find the required areas. To avoid this, one could
suggest the method of ' integration by using the concept of commulative density function as has been done in the
case of other continuous probability distributions. The CDF in this case is given by
𝑥−𝜇 2
F(x)= ∫ 1/ 2𝜋𝜎 e − dx
2𝜎 2
But unfortunately, the analytical integration of this function is not possible. One may try to integrate it numerically
but then that requires new methods with which you may not be familiar. Thus to avoid all these difficulties, we take
the help of the method of transformation of the variable. In other words, we replace the normal variable X with
Mean 𝜇 and standard deviation 𝜎 by a new variable Z with mean 0 and standard deviation 1 i.e.
N(𝜇,𝜎2 )→ N(0,1)
This can be done by means of the transformation
𝜘−𝜇
Z= 𝜎
Then, it is easy to verify that Mean of Z = 0 Variance of Z = 1 Accordingly, we define a new normal variable Z and
call it standard normal variable.

Standard Normal Distribution :

A normal variable with mean zero and standard deviation 1 is called a standard normal variable. The distribution
of this variable is called a standard normal distribution. Its density function is given by
1
f(z)= 2𝛬ഥ 𝑒^[ − 1Τ2 𝑧2], -∞<z< ∞
The graph of this standard normal distribution is shown in the Figure

The commulative distribution function (cdf). of Z is a -n by φ (z) where

𝑧
1
φ (z)= P(Z≤z)=ഥ 2𝜋
𝑒^[ − 1Τ2 𝑧2] dz
−∞
Note that we Use the special notation 4 (z) instead of the usual F(Z) for the normal distribution function. it is not
possible to evaluate this integrd.by the methods of integration we have learnt. However,, numerical
approximations for integral of this type can be done. The same have been tabulated and are given in able ' 1'.
The bell-shaped curve in Fig. represents the graph of the PDF of Z and the shaded area represents φ (z). Because
of the symmetry of the standard normal PDF, we have φ (-z)=1-φ (z) ∀ z Real.
In view of this, in the table below we list only those probabilities φ (z) associated with non-negative values of Z.
We denote- P (Z ≥ b) = P (Z > b) = φ (b), P (Z ≥ c) = φ (c) = P (Z < c),
P(a ≤ Z ≤ b) = φ (b) - φ (a), P (a ≥ Z ≥ b) = φ (a) - φ (b)
 CONCLUSION :

The application of continuous probability distributions is critical across various fields for modeling and analyzing real-
world phenomena where outcomes are continuous in nature. By providing tools for describing and predicting the
likelihood of events within a continuous range, these distributions, such as the normal, exponential, and uniform
distributions, enable robust decision-making in areas such as engineering, finance, science, and operations research.
They help in understanding uncertainty, estimating risks, and optimizing processes. Overall, continuous probability
distributions play a vital role in solving complex problems by offering a mathematical foundation to handle uncertainty
and variability.

APPLICATION IN ENGINEERING FIELD :

Continuous probability distributions have a wide range of applications in the engineering field, providing tools to model
uncertainties and optimize performance in various systems. Some key applications include:
1. Reliability Engineering: Exponential and Weibull distributions are commonly used to model the time between failures
of components and systems, enabling engineers to predict system reliability and maintenance schedules.
2. Signal Processing: Gaussian (normal) distribution is applied in noise modeling for signals, helping in the design of filters
and optimizing signal-to-noise ratios in communications systems and sensor data processing.
3. Quality Control: Continuous probability distributions, particularly the normal distribution, are used to assess variations
in manufacturing processes, enabling engineers to monitor product quality and maintain process control through
statistical process control (SPC).
4. Structural Engineering: Probability distributions help in modeling uncertainties in material properties, loads, and
environmental factors, leading to safer and more efficient design of structures such as bridges, buildings, and dams.
5. Operations Research and Optimization: Engineers use continuous probability distributions to model demand, supply,
and operational uncertainties in logistics and supply chain management, allowing for better optimization of resources
and cost-effective decision-making.
6. Thermal and Fluid Dynamics: Continuous distributions are used to model temperature fluctuations, fluid flow, and
heat transfer, allowing engineers to optimize the design and performance of heating, ventilation, and air conditioning
(HVAC) systems, engines, and power plants.
7. Renewable Energy Systems: Continuous probability distributions model wind speeds, solar irradiance, and other
environmental factors, helping engineers optimize the design and placement of wind turbines and solar panels for
maximum energy efficiency.
These applications demonstrate the crucial role continuous probability distributions play in managing uncertainty,
optimizing performance, and ensuring reliability in engineering projects.
 RESOURCES/BIBLIOGRAPHY:

Here are some key resources for studying and understanding the application of continuous probability
distributions, especially in engineering and related fields:
1. Books
• "Probability and Statistics for Engineers and Scientists" by Ronald E. Walpole and Raymond H. Myers
• This textbook provides a comprehensive foundation in probability theory and statistical
analysis, with real-world engineering applications of continuous probability distributions.
• "Introduction to Probability Models" by Sheldon M. Ross
• A well-known resource that covers both discrete and continuous probability models, with
examples in reliability, queuing theory, and other engineering fields.
• "Applied Statistics and Probability for Engineers" by Douglas C. Montgomery and George C. Runger
• This book includes applications of continuous distributions in engineering problems, with a
focus on quality control, reliability, and process optimization.
• "Probability Theory: The Logic of Science" by E. T. Jaynes
• A resource for understanding the principles of probability theory, including continuous
distributions, with applications to engineering and science.