Probability

IPM – Term II, January 2023

Dr. Landis Conrad Felix Michel

6.1
Probability Distributions for Continuous Random
Variables
Probability Distributions for Continuous Random Variables

When a random variable x is discrete, you can assign a

positive probability to each value that x can take and get the
probability distribution for x. The sum of all the probabilities
associated with the different values of x is 1. However, not
all experiments result in random variables that are discrete.

Continuous random variables, such as heights and

weights, length of life of a particular product, or
experimental laboratory error, can assume the infinitely
many values corresponding to points on a line interval.

3
Probability Distributions for Continuous Random Variables

The outline of the histogram will change slightly, for the most
part becoming less and less irregular, as shown in figure.

Relative frequency histograms for increasingly large sample sizes

Figure 6.1

• As the number of measurements becomes very large and the class widths become very narrow, the
relative frequency histogram appears more and more like the smooth curve shown in Figure 6.1(d).
This smooth curve describes the probability distribution of the continuous random variable.

4
Probability Distributions for Continuous Random Variables

The depth or density of the probability, which varies with x, may be described by a
mathematical formula f (x), called the probability distribution or probability density
function for the random variable x.

• The area under a continuous

probability distribution is equal
to 1.

• The probability that x will fall into a particular interval—say, from a to b—is equal
to the area under the curve between the two points a and b.

5
The Continuous Uniform Probability Distribution

The continuous uniform random variable is used to model the

behavior of a random variable whose values are uniformly or evenly
distributed over a given interval.

If we describe this interval in general as an interval from a to b, the

formula or probability density function that describes this random
variable x is given by

6
The Continuous Uniform Probability Distribution

A graph of this probability distribution is shown in figure.

The continuous uniform probability distribution

Figure 6.3

7
The Continuous Uniform Probability Distribution

The total area under the probability distribution is

and the probability that x lies in a particular
interval can be calculated as the area under the rectangle
over that area.

Finally, the mean and variance of x are given by

8
Example 6.1
The error introduced by rounding an observation to the
nearest inch has a uniform distribution over the interval from
−.5 to .5. What is the probability that the rounding error is
less than .2 in absolute value?

Figure 6.4

9
Example 6.1 – Solution
This probability corresponds to the area under the distribution
between x = −.2 and x = .2, as shown in Figure 6.4. Since the
height of the rectangle is 1, the area of the rectangle from −.2
< x < .2 is:

P(−.2 < x < .2) = [.2 − (−.2)] × 1 = .4

10
The Exponential Probability Distribution

The exponential random variable is used to model

positive continuous random variables such as waiting times
or lifetimes associated with electronic components. The
probability density function is given by

and is 0 otherwise. The parameter λ (the Greek letter

“lambda”) is often referred to as the intensity and is related
to the mean and variance as

so that μ = σ.

11
The Exponential Probability Distribution

A graph of an exponential distribution is shown in figure.

An exponential probability distribution

Figure 6.5

To find areas under this curve, you can use the fact that
to calculate right-tailed probabilities.
The left-tailed probabilities can be calculated using the
complement rule as

12
Example 6.2
The waiting time at a supermarket checkout counter has an
exponential distribution with an average waiting time of 5
minutes. What is the probability that you will have to wait
more than 10 minutes at the checkout counter?

Figure 6.6

13
Example 6.2 – Solution
Since the average waiting time at the checkout counter is
μ = 5 minutes and because μ = 1/λ, we can find 5 = 1/λ or
λ = .2.

The probability density function then is given by

(shown in Figure 6.6) and the probability
to be calculated is the shaded area in the figure.

Use the general formula for P(x > a) to find

14
The Normal Probability Distribution

A large number of random variables observed in nature

possess a frequency distribution that is approximately
mound-shaped and can be modelled by a normal
probability distribution.

15
The Normal Probability Distribution

The formula or probability density function (pdf) that

generates this distribution is shown below
Normal Probability Distribution

The symbols e and π are mathematical constants given

approximately by 2.7183 and 3.1416, respectively; μ and σ
(σ > 0) are parameters that represent the population mean
and standard deviation, respectively.

16
The Normal Probability Distribution

The graph of a normal probability distribution with mean μ

and standard deviation σ is shown in figure.

Normal probability distribution

Figure 6.7

17
The Normal Probability Distribution

The mean μ locates the center of the distribution, and the

distribution is symmetric about its mean μ.

Since the total area under the normal probability distribution is

equal to 1, this symmetry implies that the area to the right of μ is .5
and the area to the left of μ is also .5.

18
The Normal Probability Distribution
The shape of the distribution is determined by σ, the population
standard deviation. Figure shows three normal probability distributions
with different means and standard deviations.

• Notice the differences in shape

and location. Large values of σ
reduce the height of the curve and
increase the spread; small values
of σ increase the height of the
curve and reduce the spread.

Normal probability distributions with differing values of μ and σ

19
The Standard Normal Random Variable (1 of 5)

To standardize a normal random variable x, we express its

value as the number of standard deviations (σ) it lies to the
left or right of its mean μ.

The standardized normal random variable, z, is defined as

or equivalently,
x = μ + zσ

This is the familiar z-score, used to detect outliers.

20
The Standard Normal Random Variable (2 of 5)

Look at the formula, z = (x − μ) / σ :

• When x is less than the mean μ, the value of z is

negative.
• When x is greater than the mean μ, the value of z is
positive.
• When x = μ, the value of z = 0.

21
The Standard Normal Random Variable (3 of 5)

The probability distribution for z, shown in figure, is called

the standardized normal distribution because its mean is
0 and its standard deviation is 1.

Standardized normal distribution

Figure 6.9

22
Example 6.5
Find P(−.5 ≤ z ≤ 1.0). This probability is the area between
z = −.5 and z = 1.0, as shown in Figure.

Area under the standard normal curve for Example 6.5

Figure 6.12

23
Example 6.5 – Solution
To find the area marked A2, we subtract the two entries:

A2 = (A1 + A2) − A1 = .8413 − .3085 = .5328

That is, P(−.5 ≤ z ≤ 1.0) = .5328.

24
Example 6.6
Find the probability that a normally distributed random
variable will fall within these ranges:

1. One standard deviation of its mean

2. Two standard deviations of its mean

25
Example 6.6
1. Since the standard normal random variable z measures
the distance from the mean in units of standard
deviations, you need to find

P(−1 ≤ z ≤ 1) = .8413 − .1587 = .6826

2. As in part 1, P(−2 ≤ z ≤ 2) = .9772 − .0228 = .9544.

These probabilities agree with the approximate values of

68% and 95% in the Empirical Rule.

26
Example 6.7
Find the value of z—say c—such that .95 of the area is
within ±c standard deviations of the mean.

27
Example 6.7 – Solution
The shaded area in figure is the area within ± c standard
deviations of the mean, which needs to be equal to .95.

Figure 6.13

28
Example 6.7 – Solution
The “tail areas” under the curve are not shaded, and have
a combined area of 1 − .95 = .05.

Because of the symmetry of the normal curve, these two

tail areas have the same area, so that A1 = .05 ∕ 2 = .025 in
Figure 6.13.

Thus, the entire cumulative area to the left of −c equals

A1 = .025.

29
Example 6.7 – Solution
This area can be calculated with the inverse of the
Cumulative Distribution Function, which for each A,
calculates the value of Z.

Hence, −c = −1.96 or c = 1.96. Note that this result is very

close to the approximate value, z = 2, used in the Empirical
Rule.

A common notation for a value of z having area .025 to its

right is z.025 = 1.96.

30
Example 6.8
Let x be a normally distributed random variable with a mean
of 10 and a standard deviation of 2. Find the probability that
x lies between 11 and 13.6.

Solution:
Consider the inequality 11 ≤ x ≤ 13.6. As long as we subtract
the same number across the inequality or multiply/divide by
the same positive number, the inequality will remain the
same.

31
Example 6.8 – Solution
For this interval

11 ≤ x ≤ 13.6

11 − 10 ≤ x − 10 ≤ 13.6 − 10

.5 ≤ z ≤ 1.8

32
Example 6.8 – Solution
The desired probability is therefore
P(11 ≤ x ≤ 13.6) = P(.5 ≤ z ≤ 1.8),

the area lying between z = .5 and z = 1.8, as shown in figure.

Figure 6.14

33
The Normal Approximation to the Binomial Probability Distribution

The Normal Approximation to the Binomial Probability

Distribution
Let x be a binomial random variable with n trials and
probability p of success. The probability distribution of x is
approximated using a normal curve with

This approximation is adequate as long as n is large and p

is not too close to 0 or 1.

34
The Normal Approximation to the Binomial Probability Distribution

Remember that the bars of the binomial histogram have a

width of 1. Therefore, the area under a single bar, say at x = a,
is equal to P(x = a).

Hence, it is possible to approximate a binomial probability (the

area under the bars) using the area under an appropriate
normal curve over the same region.

35
The Normal Approximation to the Binomial Probability Distribution

Figures show the binomial probability histograms for n = 25

with p = .5 and p = .1, respectively. The distribution in Figure
6.17 is exactly symmetric.

The binomial probability distribution for n = 25 The binomial probability distribution and the
and p = .5 and the approximating normal approximating normal distribution for n = 25
distribution with μ = 12.5 and σ = 2.5 and p = .1
Figure 6.17 Figure 6.18

36
Example 6.11 – Solution
To use the normal approximation, first find the appropriate
mean and standard deviation for the normal curve:

The probability that you need corresponds to the area of the

three rectangles lying over x = 8, 9, and 10.

The equivalent area under the normal curve lies between

x = 7.5 (the lower edge of the rectangle for x = 8) and
x = 10.5 (the upper edge of the rectangle for x = 10).

37
Example 6.11
Use the normal curve to approximate the probability that x
= 8, 9, or 10 for a binomial random variable with n = 25 and
p = .5. Compare this approximation to the exact binomial
probability.

Solution:

P(x = 8, 9, or 10) = P(x ≤ 10) − P(x ≤ 7) =.212 − .022 = .190

38
Example 6.11 – Solution
This area is shaded in figure.

The binomial probability distribution for n = 25

and p = .5 and the approximating normal
distribution with μ = 12.5 and σ = 2.5
Figure 6.17

39
The Normal Approximation to the Binomial Probability Distribution (5 of 7)

The normal approximation works well when the binomial

histogram is roughly symmetric. This happens when the
binomial distribution is not “piled up” near 0 or n—that is,
when it can spread out at least two standard deviations from
its mean without exceeding its limits, 0 and n. This concept
leads us to a simple rule of thumb:

Rule of Thumb
The normal approximation to the binomial probabilities
will be adequate if both
np > 5 and nq > 5 np>5 and nq>5 means that curve is symmetrical
normal approximation is always symmetrical

40
References – Additional Readings
• Chapters 6, “Introduction to Probability and Statistics”, 2020, William Mendenhall, Robert J.
Beaver, Barbara M. Beaver, 15TH Edition, Cengage Learning, ISBN: 1337554421

•.