McGraw-Hill/Irwin Copyright 2009 by The McGraw-Hill Companies, Inc. All Rights Reserved.

Confidence Intervals
Chapter 8
8-2
Confidence Intervals
8.1 z-Based Confidence Intervals for a
Population Mean: Known
8.2 t-Based Confidence Intervals for a
Population Mean: Unknown
8.3 Sample Size Determination
8.4 Confidence Intervals for a Population
Proportion
8.5 Confidence Intervals for Parameters of
Finite Populations (Optional)
8.6 A Comparison of Confidence Intervals
and Tolerance Intervals (Optional)
8-3
z-Based Confidence Intervals for a
Mean: Known
The starting point is the sampling
distribution of the sample mean
Recall that if a population is normally
distributed with mean and standard
deviation , then the sampling distribution
of x is normal with mean
x
= and
standard deviation
Use a normal curve as a model of the
sampling distribution of the sample mean
Exactly, because the population is normal
Approximately, by the Central Limit Theorem
for large samples
n
x
o = o
8-4
The Empirical Rule
68.26% of all possible sample means
are within one standard deviation of the
population mean
95.44% of all possible observed values
of x are within two standard deviations
of the population mean
99.73% of all possible observed values
of x are within three standard deviations
of the population mean
8-5
Example 8.1: The Car Mileage Case
Assume a sample size (n) of 5
Assume the population of all individual
car mileages is normally distributed
Assume the population standard
deviation () is 0.8


The probability that x with be within 0.7
of is 0.9544
358 . 0
5
8 . 0
= = =
n
x
o
o
8-6
Example 8.1 The Car Mileage Case
Continued
Assume the sample mean is 31.3
That gives us an interval of [31.3 0.7]
= [30.6, 32.0]
The probability is 0.9544 that the
interval [x 2] contains the population
mean

8-7
Example 8.1: The Car Mileage Case
#3
That the sample mean is within 0.7155 of
is equivalent to
x will be such that the interval [x 0.7115]
contains
Then there is a 0.9544 probability that x
will be a value so that interval [x 0.7115]
contains
In other words
P(x 0.7155 x + 0.7155) = 0.9544
The interval [x 0.7115] is referred to as
the 95.44% confidence interval for
8-8
Example 8.1: The Car Mileage Case
#4
Three 95.44% Confidence Intervals for
Thee intervals
shown
Two contain
One does not
8-9
Example 8.1: The Car Mileage Case
#5
According to the 95.44% confidence interval,
we know before we sample that of all the
possible samples that could be selected
There is 95.44% probability the sample mean
is such the interval [x 0.7155] will contain
the actual (but unknown) population mean
In other words, of all possible sample means,
95.44% of all the resulting intervals will contain the
population mean
Note that there is a 4.56% probability that the
interval does not contain
The sample mean is either too high or too low
8-10
Generalizing
In the example, we found the probability that
is contained in an interval of integer
multiples of o
x
More usual to specify the (integer) probability
and find the corresponding number of o
x

The probability that the confidence interval
will not contain the population mean is
denoted by o
In the mileage example, o = 0.0456
8-11
Generalizing Continued
The probability that the confidence interval
will contain the population mean is
denoted by 1 o
1 o is referred to as the confidence
coefficient
(1 o) 100% is called the confidence level
Usual to use two decimal point probabilities
for 1 o
Here, focus on 1 o = 0.95 or 0.99
8-12
General Confidence Interval
In general, the probability is 1 o that the
population mean is contained in the interval


The normal point z
o/2
gives a right hand tail area
under the standard normal curve equal to o/2
The normal point - z
o/2
gives a left hand tail area
under the standard normal curve equal to o/2
The area under the standard normal curve
between -z
o/2
and z
o/2
is 1 o
| |
(

o
= o
o o
n
z x z x
x 2 2
8-13
Sampling Distribution Of All Possible
Sample Means
8-14
z-Based Confidence Intervals for a
Mean with Known
If a population has standard deviation o
(known),
and if the population is normal or if
sample size is large (n > 30), then
a (1-o)100% confidence interval
for is
(

o
+
o
=
o

o o o
n
z x ,
n
z x
n
z x
2 2 2
8-15
95% Confidence Level
For a 95% confidence level, 1 o = 0.95, so
o = 0.05, and o/2 = 0.025
Need the normal point z
0.025
The area under the standard normal curve
between -z
0.025
and z
0.025
is 0.95
Then the area under the standard normal curve
between 0 and z
0.025
is 0.475
From the standard normal table, the area is 0.475
for z = 1.96
Then z
0.025
= 1.96
8-16
95% Confidence Interval
The 95% confidence interval is
| |
(

o
+
o
=
(

o
= o
n
. x ,
n
. x
n
. x z x
x .
96 1 96 1
96 1
025 0
8-17
99% Confidence Interval
For 99% confidence, need the normal
point z
0.005
Reading between table entries in the
standard normal table, the area is 0.495 for
z
0.005
= 2.575
The 99% confidence interval is
| |
(

o
+
o
=
(

o
= o
n
. x ,
n
. x
n
. x z x
x .
575 2 575 2
575 2
025 0
8-18
The Effect of a on Confidence Interval
Width
z
o/2
= z
0.025
= 1.96

z
o/2
= z
0.005
= 2.575

8-19
Example 8.2: Car Mileage Case
Given
x = 31.56
= 0.8
n = 50
Will calculate
95 Percent confidence interval
99 Percent confidence interval
8-20
Example 8.2: 95 Percent Confidence
Interval
| | 78 . 31 , 34 . 31
222 . 0 56 . 31
50
8 . 0
96 . 1 56 . 31
025 . 0
=
=
=
n
z x
o
8-21
Example 8.2: 99 Percent Confidence
Interval
| | 85 . 31 , 27 . 31
294 . 0 56 . 31
50
8 . 0
575 . 2 56 . 31
005 . 0
=
=
=
n
z x
o
8-22
Notes on the Example
The 99% confidence interval is slightly wider
than the 95% confidence interval
The higher the confidence level, the wider the
interval
Reasoning from the intervals:
The target mean should be at least 31 mpg
Both confidence intervals exceed this target
According to the 95% confidence interval, we can
be 95% confident that the mileage is between
31.33 and 31.78 mpg
We can be 95% confident that, on average, the mean
exceeds the target by at least 0.33 and at most 0.78 mpg
8-23
t-Based Confidence Intervals for a
Mean: o Unknown
If o is unknown (which is usually the
case), we can construct a confidence
interval for based on the sampling
distribution of



If the population is normal, then for any
sample size n, this sampling distribution
is called the t distribution
n s
x
t

=
8-24
The t Distribution
The curve of the t distribution is similar to
that of the standard normal curve
Symmetrical and bell-shaped
The t distribution is more spread out than
the standard normal distribution
The spread of the t is given by the number
of degrees of freedom
Denoted by df
For a sample of size n, there are one fewer
degrees of freedom, that is,
df = n 1
8-25
Degrees of Freedom and the
t-Distribution
As the number of degrees of freedom increases, the spread
of the t distribution decreases and the t curve approaches
the standard normal curve
8-26
The t Distribution and Degrees of
Freedom
As the sample size n increases, the
degrees of freedom also increases
As the degrees of freedom increase, the
spread of the t curve decreases
As the degrees of freedom increases
indefinitely, the t curve approaches the
standard normal curve
If n 30, so df = n 1 29, the t curve is
very similar to the standard normal curve
8-27
t and Right Hand Tail Areas
Use a t point denoted by t
o
t
o
is the point on the horizontal axis under
the t curve that gives a right hand tail equal
to a
So the value of t
o
in a particular situation
depends on the right hand tail area a and
the number of degrees of freedom
df = n 1
o = 1 a , where 1 a is the specified
confidence coefficient
8-28
t and Right Hand Tail Areas
8-29
Using the t Distribution Table
Rows correspond to the different values of df
Columns correspond to different values of a
See Table 8.3, Tables A.4 and A.20 in
Appendix A and the table on the inside cover
Table 8.3 and A.4 gives t points for df 1 to 30, then
for df = 40, 60, 120 and
On the row for , the t points are the z points
Table A.20 gives t points for df from 1 to 100
For df greater than 100, t points can be approximated by
the corresponding z points on the bottom row for df =
Always look at the accompanying figure for
guidance on how to use the table
8-30
Using the t Distribution
Example: Find t
o
for a sample of size
n=15 and right hand tail area of 0.025
For n = 15, df = 14
o = 0.025
Note that a = 0.025 corresponds to a
confidence level of 0.95
In Table 8.3, along row labeled 14 and
under column labeled 0.025, read a table
entry of 2.145
So t
o
= 2.145
8-31
Using the t Distribution Continued
8-32
t-Based Confidence Intervals for a
Mean: o Unknown
If the sampled population is normally
distributed with mean , then a (1o)100%
confidence interval for is





t
o/2
is the t point giving a right-hand tail area
of o/2 under the t curve having n1 degrees of
freedom
n
s
t x
2 o

8-33
Example 8.4 Debt-to-Equity Ratios
Estimate the mean debt-to-equity ratio of the
loan portfolio of a bank
Select a random sample of 15 commercial
loan accounts
x = 1.34
s = 0.192
n = 15
Want a 95% confidence interval for the ratio
Assume all ratios are normally distributed but
unknown
8-34
Example 8.4 Debt-to-Equity Ratios
Continued
Have to use the t distribution
At 95% confidence, 1 o = 0.95 so o = 0.05
and o/2 = 0.025
For n = 15, df = 15 1 = 14
Use the t table to find t
o
/2 for df = 14, t
o
/2 =
t
0.025
= 2.145
The 95% confidence interval:
| | 4497 . 1 , 2369 . 1 1064 . 0 3433 . 1
15
192 . 0
145 . 2 343 . 1
025 . 0
= =
=
n
s
t x
8-35
Sample Size Determination (z)
If o is known, then a sample of size






so that x is within B units of , with
100(1-o)% confidence
2
2
|
|
.
|

\
|
o
=
o
B
z
n
8-36
Sample Size Determination (t)
If is unknown and is estimated from s,
then a sample of size




so that x is within B units of , with
100(1-o)% confidence. The number of
degrees of freedom for the t
o/2
point is
the size of the preliminary sample
minus 1
2
2
|
|
.
|

\
|
=
B
s t
n
o
8-37
Example 8.6: The Car Mileage Case
What sample size is needed to make
the margin of error for a 95 percent
confidence interval equal to 0.3?
Assume we do not know
Take prior example as preliminary
sample
s = 0.7583
z
o/2
= z
0.025
= 1.96
t
o/2
= t
0.025
= 2.776 based on n-1 = 4 d.f.
8-38
Example 8.6: The Car Mileage Case
Continued
( )
50 24 . 49
3 . 0
7583 . 0 776 . 2
2 2
2 /
=
|
.
|

\
|
=
|
.
|

\
|
=
E
s t
n
o
8-39
Example 8.7: The Car Mileage Case
Want to see that the sample of 50 mileages
produced a 95 percent confidence interval
with a margin or error of 0.3






The margin of error is 0.227, which is smaller
than the 0.3 desired
| | | | 79 . 31 , 33 . 31 227 . 0 56 . 31
50
798 . 0
010 . 2 56 . 31
025 . 0
= =
=
(

=
(

n
s
t x
8-40
Confidence Intervals for a Population
Proportion
If the sample size n is large, then a (1-
o)100% confidence interval for is




Here, n should be considered large if
both
n p 5
n (1 p ) 5
( )
n
p

z p

o
1
2
8-41
Example 8.8: The Cheese Spread Case
Would like a confidence interval for ,
the proportion of purchases who would
stop buying if new spout used
Wish to see if n=1,000 is large enough
Point estimate p is 0.63
np = 1,000(0.063) = 63 > 5
n(1-p) = 1,000(0.937) = 937 > 5

8-42
Example 8.8: The Cheese Spread Case
Continued




Upper limit is less than 0.10
Strong evidence that true proportion is
less than 0.10
( ) ( )( )
| |
| | 0781 , 0 , 0479 . 0
0151 . 0 063 . 0
000 , 1
937 . 0 063 . 0
96 . 1 063 . 0

025 . 0
=
=
(

=
(

n
p p
z p
8-43
Determining Sample Size for Confidence
Interval for
A sample size given by the formula




will yield an estimate p, precisely within B
units of , with 100(1-o)% confidence.

Note that the formula requires a preliminary
estimate of p. The conservative value of
p=0.5 is generally used when there is no prior
information on p.
( )
2
2
1
|
|
.
|

\
|
=
o
B
z
p p n
8-44
Confidence Intervals for Population
Mean and Total for a Finite Population
For a large (n 30) random sample of
measurements selected without replacement
from a population of size N, a (1- o)100%
confidence interval for is



A (1- o)100% confidence interval for the
population total is found by multiplying the
lower and upper limits of the corresponding
interval for by N
N
n N
n
s
z x

o 2
8-45
Confidence Intervals for Proportion
and Total for a Finite Population
For a large random sample of measurements
selected without replacement from a
population of size N, a (1- o)100%
confidence interval for is




A (1- o)100% confidence interval for the total
number of units in a category is found by
multiplying the lower and upper limits of the
corresponding interval for by N
( )
|
.
|

\
|

N
n N
n
p

1
1
8-46
A Comparison of Confidence Intervals
and Tolerance Intervals
A tolerance interval contains a specified
percentage of individual population
measurements
Often 68.26%, 95.44%, 99.73%
A confidence interval is an interval that
contains the population mean , and the
confidence level expresses how sure we are
that this interval contains
Often confidence level is set high (e.g., 95% or
99%)
Such a level is considered high enough to provide
convincing evidence about the value of
8-47
Example 8.15 Car Mileage Case
Tolerance intervals shown:
[x s] contains 68% of all individual cars
[x 2s] contains 95.44% of all individual cars
[x 3s] contains 99.73% of all individual cars
The t-based 95%
confidence interval for is
[31.32, 31.78], so we can
be 95% confident that
lies between 31.23 and
31.78 mpg
8-48
Selecting an Appropriate Confidence
Interval for a Population Mean