Vous êtes sur la page 1sur 40

STATISTICS-I

COURSE INSTRUCTOR:
TEHSEEN IMRAAN

CHAPTER 7
CONTINUOUS
PROBABILITY
DISTRIBUTION

INTRODUCTION
This chapter continues our study of
probability distributions by examining
the continuous probability distribution.
Recall that a continuous probability
distribution can assume an infinite
number of values within a given range.
As an example, the weights for a
sample of small engine blocks are:
54.3, 52.7, 53.1 and 53.9 pounds.

INTRODUCTION
We consider two families of
continuous probability distributions,
the uniform probability
distribution and the normal
probability distribution. Both of
these distributions describe the
likelihood of a continuous random
variable that has an infinite number
of possible values within a specified
range.

INTRODUCTION
An example of a uniform probability
distribution is the flight time between
NY and Chicago. Suppose the time to fly
from NY to Chicago is uniformly
distributed within a range of 55 minutes
to 75 minutes. We can determine the
probability that we can fly from NY to
Chicago in less than 60 minutes. Flight
time is measured on a continuous scale.

INTRODUCTION
The normal probability distribution is
described by its mean and standard
deviation. Suppose the life of an
automobile battery follows the
normal distribution with a mean of 36
months and a standard deviation of 3
months. We can determine the
probability that a battery will last
between 36 and forty months. Life of
a battery is measured on a

Characteristics of the Uniform Probability Distribution

A continuous probability distribution with its


values spread evenly over a range of values
that are rectangular in shape and are defined
by minimum and maximum values.
A uniform distribution is shown in Chart 7-1.
The distributions shape is rectangular and
has a minimum value of "a" and a maximum
value of "b". The height of the distribution is
uniform for all values between "a " and "b".
This implies that all the values in the range
are equally likely.

CHART 7-1

Mean of a Uniform
Distribution

Mean of a Uniform
Distribution
The mean of a uniform distribution is
located in the middle of the interval
between the minimum value of "a " and
a maximum value of "b".
For example: Suppose that the time to
fly from NY to Chicago is uniformly
distributed within a range of 55 minutes
minimum to 75 minutes maximum. The
mean is found by using Formula [71]:

Mean of a Uniform
Distribution

Standard Deviation of a Uniform


Distribution

For the flight time from NY to Chicago


example the standard deviation is
calculated
using
Formula
[72]

Another key element of the uniform distribution is the height, P(x). The height is the
same for all values of the random variable "x ". It is calculated using Formula[73]:

In Chapter 6, we discussed the fact


that probability distributions are
useful when making probability
statements concerning the values of
a random variable. Also for
continuous random variables, areas
within the distribution represent
probabilities. Recall that:

FOR ALL VALUES OF x

The relationship between area and probabilities is applied to the uniform distribution and its rectangular
shape using the area of a rectangle formula. Recall that:

Thus for any uniform distribution, the area under the curve is always 1.
For the flight time from NY to Chicago example the area is:

Normal probability
distribution
A continuous probability distribution
uniquely determined by m and s.
The major characteristics of the
normal distribution are:
The normal distribution is "bell-shaped" and
the mean, median, and mode are all equal
and are located in the center of the
distribution. Exactly one-half of the area
under the normal curve is above the center
and one-half of the area is below the center.

Normal probability
distribution
The distribution is symmetrical about the mean.
A vertical line drawn at the mean divides the
distribution into two equal halves and these
halves possess exactly the same shape.
It is asymptotic. That is, the tails of the curve
approach the X-axis but never actually touch it.
A normal distribution is completely described
by its mean and standard deviation. This
indicates that if the mean and standard
deviation are known, a normal distribution can
be constructed and its curve drawn.

Normal probability
distribution
There is a "family" of normal probability
distributions. This means there is a
different normal distribution for each
combination of m and s.

These characteristics are summarized in the graph.

The Standard Normal Probability Distribution

Standard normal distribution: A


normal distribution with a mean of 0
and a standard deviation of 1.
z value: The signed distance
between a selected value designated
X, and the population mean, m ,
divided by the population standard
deviation, s .

The formula for a specific standardized z value is text formula [75]:

Where:

X is the value of any particular


observation or measurement.
m is the mean of the distribution.
s is the standard deviation of the
distribution.
z is the standardized normal value,
usually called the z value.

Applications of the Standard Normal Distribution

To obtain the probability of a value


falling in the interval between the
variable of interest (X) and the mean
(m ), we first compute the distance
between the value (X) and the mean
(m ). Then we express that difference
in units of the standard deviation by
dividing (X - m ) by the standard
deviation. This process is called
standardizing.

To illustrate the probability of a value


being between a selected X value
and the mean m , suppose the mean
useful life of a car battery is 36
months, with a standard deviation of
3 months. What is the probability
that such a battery will last between
36 and 40 months?

The first step is to convert the 40 months to an


equivalent standard normal value, using formula [7
5]. The computation is:

a table for the areas under the


normal curve
z

0.00

0.01

0.02

0.03

0.04

0.05

1.0

1.1

0.3665

0.3686

0.3708

0.3729

1.2

0.3869

0.3888

0.3907

0.3925

1.3

0.4049

0.4066

0.4082

0.4099

1.4

0.4207

0.4222

0.4236

0.4251

To use the table, the z value of 1.33 is split


into two parts, 1.3 and 0.03. To obtain the
probability go down the left-hand column to
1.3, then move over to the column headed
0.03 and read the probability. It is 0.4082.
The probability that a battery will last
between 36 and 40 months is 0.4082.
Other probabilities may be calculated, such
as more than 46 months, and less than 33
months.

Empirical Rule
About 68 percent of the area under the normal
curve is within plus one and minus one
standard deviation of the mean. This can be
written as m 1s.
About 95 percent of the area under the normal
curve is within plus and minus two standard
deviations of the mean, written m 2s.
Practically all of the area under the normal
curve is within three standard deviations of the
mean, written m 3s.

Empirical Rule

The Normal Approximation to


the Binomial
A binomial probability can be estimated
using the normal distribution.
To apply the normal approximation to
the binomial, both np and nq must be
greater than 5. The sample size, or the
number of trials, is designated by n,
and p is the probability of a success.
The mean and the standard deviation
of the binomial are computed by:

Example
To illustrate, suppose 60 percent of
the applications for an exclusive
credit card are approved. In a sample
of 200 applications, what is the
probability that 130 or more
applications are approved?
First verify that both n p and nq
exceed 5. For n = 200 and p = 0.6

Thus the normal approximation to


the binomial may be used.
The mean and standard deviation
are computed as follows:

This distribution is standardized by


formula [75] and m = 120, s =
6.93 and letting X =129.5 (Not 130)

Why is 129.5 used instead of 130? It is


used to "correct" for the fact that a
continuous distribution (the normal) is
used to approximate a discrete distribution
(the binomial). On a continuous scale the
value 130 would range from 129.5 to
130.5. On a discrete scale there would be a
"gap" between 129 and 130 where there
would not be any probability. The 0.50 is
called the correction for continuity.

Continuity correction
factor
The value 0.5 subtracted or added, depending
on the question, to a selected value when a
discrete probability distribution is
approximated by a continuous probability
distribution
The probability of a z value between 0 and
1.37 is 0.4147 (See Appendix D). Therefore
the probability of a z value greater than 1.37
is 0.0853, found by 0.5000 - 0.4147. So, the
probability that 130 or more applications will
be approved is 0.0853.

Vous aimerez peut-être aussi