CHAPTER

Estimation 3
Outline
1.1
3.1 Descriptive and Inferential Statistics
Introduction
3.2 Interval Estimation for a Mean
• Small and Large Sample
3.3 Interval Estimation for the Difference
Between Two Means (Independent Sample)
3.4 Interval Estimation for the Difference
Between Two Means (Dependent Sample)

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
Copyright © 2012 The McGraw-Hill Companies, Inc. Slide 1
Inferential Statistics

o The area of statistics that deals with such decision-

making procedures.
o Methods use to draw conclusions about
characteristics of population based on sample data.
o The application of analytic procedures on the sample
of population to infer (generalize) the results
obtained to the target population
1. Estimation
2. Hypothesis testing.
Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
INTRODUCTION

Methods for making inferences

Statistical
about population parameters fall
Inference
into one of the two categories:
estimation or hypothesis testing.

Estimation: Hypothesis
Estimation
Estimation is a process of testing
estimating the value of a
population parameter based on
the sample statistic.
Point Interval
Hypothesis Testing: Estimation Estimation
Hypothesis testing is a decision
making process for evaluating
claims about a population
parameter.
Estimator

The sample statistic used to estimate a population parameter is known as an estimator.

Sample Statistic Parameter
•Sample mean, x •Population mean, 
2 2
•Sample variance, s •Population variance, 

A point estimate of a population parameter is a single number obtained from the sample.

An estimate of a population parameter given by two numbers between which the parameter
may be considered to lie is called an interval estimate.

Example

To estimate µ, the population mean diameter of piston rings for a car engine produce by a
manufacturer, the diameters of a random sample of 15 piston rings were measured. The
sample mean was calculated and the value obtained is 8.23 mm.

If we estimate the mean diameter of all piston rings produced by the manufacturer as
8.23 mm we are giving the point estimate.
If we estimate the mean diameter as (8.216mm, 8.244 mm) or 8.23±0.014 we are
giving the interval estimate.
Characteristics of good estimator
a) Unbiased
b) Efficiency
c) Consistency
d) Sufficiency

The best point estimator of the population mean is the sample mean,

¯=
𝑋
∑ 𝑋𝑖
𝑛
The best point estimator of the population standard deviation is the sample
standard deviation,

√
(∑ 𝑋 )
2

∑𝑋 2
−
𝑛
𝑆=
𝑛 −1
Confidence Interval

An interval estimate also called a confidence interval is a range of values

constructed around the point estimate.

This estimate may or may not contain the value of the parameter being estimated.

The confidence level of an interval estimate of a parameter denoted by (1-α)100%

states how much confidence we have that the interval contains the true population
parameter.
1-α is the probability that the interval estimate will contain the parameter and is
called the confidence coefficient.

Lower Point Upper

Confidence Confidence
Estimate
Limit Limit

Width of confidence interval

Example
If an interval (a,b) is such that P(a<θ<b)=0.95 then (a,b) is the 95% confidence interval for θ,
0.95 is the confidence coefficient.
Confidence Interval of the Mean for a Specific a, when
 is known, for large samples, (n ≥ 30) KENA CHECK

𝑠
¯𝑥 ± 𝑧 𝛼 / 2
√𝑛
where zα/2 is the upper 100α/2 percentage point of the standard normal distribution.

The general formula for all confidence intervals is:

Point Estimate ± (Critical Value)(Standard Error Mean)

To be more confident that the interval contains the true population mean, you must make the
interval wider.
 Normal Distribution Properties
• The normal distribution curve is bell-shaped.
• The mean, median, and mode are equal and
located at the center of the distribution.
• The normal distribution curve is unimodal (i.e.,
it has only one mode).
• The curve is symmetrical about the mean,
which is equivalent to saying that its shape is
the same on both sides of a vertical line passing
through the center.
Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Normal Distribution Properties
• The curve is continuous—i.e., there are no gaps
or holes. For each value of X, there is a
corresponding value of Y.
• The curve never touches the x-axis.
Theoretically, no matter how far in either
direction the curve extends, it never meets the
x-axis—but it gets increasingly close.
• The total area under the normal distribution
curve is equal to 1.00 or 100%.
Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
Chapter 3: Estimation Topic 3.3 : Confidence Interval for One population mean Leave blank

Example :
Using
Table 4,
find z0.025..

z0.025
=1.9600
95% Confidence Interval of the Mean

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
95% Confidence Interval of the Mean

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Example:
A random sample of n = 50 males showed a mean average daily
intake of dairy products equal to 756 grams with a standard
deviation of 35 grams. Find a 95% confidence interval for the
population average m
1.96

𝑠
¯𝑥 ± 𝑧 0.05 / 2
√𝑛
35
⇒ 7 56 ± 1.96
⇒ 7 56 ± 9.70 √ 50

746.30 <𝜇<7 65.70 grams.

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Example : Days to Sell an Aveo
A researcher wishes to estimate the number of days it takes an
automobile dealer to sell a Chevrolet Aveo. A sample of 50 cars had a
mean time on the dealer’s lot of 54 days. Assume the population
standard deviation to be 6.0 days. Find the 95% confidence interval of
the population mean.

6.0
54 ± 𝑧 0.025
√6.0
50
54 ± (1.96)
√ 50
54 ± 1.7
52.3< 𝜇<55.7
We can say with 95% confidence that the mean number of days it takes an
automobile dealer to sell a Chevrolet Aveo is between 52 and 56 days.

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Example: Waiting Times
A large department store found that it averages 362 customers per hour.
Assume that the standard deviation is 29.6 and a random sample of 40
hours was used to determine the average. Find the 99% confidence
interval of the population mean.

29.6
362 ± 𝑧 0.005
√ 40
29.6
362 ±(2.5758)
√ 40
362 ± 12.1
(350 ,374 )
Hence, one can be 99% confident that the mean number of customers that the
store averages is between 350 and 374 customers per hour.
Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Example: Credit Union Assets
The following data represent a sample of the assets (in millions
of dollars) of 30 credit unions in southwestern Pennsylvania.
Find the 90% confidence interval of the mean.
12.23 16.56 4.39
2.89 1.24 2.17
13.19 9.16 1.42
73.25 1.91 14.64
11.59 6.69 1.06
8.74 3.17 18.13
7.92 4.78 16.85
40.22 2.42 21.58
5.01 1.47 12.24
2.27 12.77 2.76

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Example: Credit Union Assets
Step 1: Find the mean and standard deviation. Using calculator, we
find =𝑋 ¯11.091 and s = 14.405.

Step 2: Substitute in the formula.

14.405
11.091 ± 𝑧 0.05
√ 30
14.405
11.091 ±(1.6449)
√30
11.091 ± 4.326
6.752<𝜇 <15.430
One can be 90% confident that the population mean of the assets of
all credit unions is between $6.752 million and $15.430 million.
Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Confidence Intervals for the Mean
When  Is Unknown and n < 30
The value of , when it is not known, must be estimated by
using s, the standard deviation of the sample.

When s is used, especially when the sample size is small

(less than 30), critical values greater than the values for z 2
are used in confidence intervals in order to keep the interval at
a given level, such as the 95%.

These values are taken from the Student t distribution, most

often called the t distribution.
Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
20
 Characteristics of the t Distribution
The t distribution is similar to the standard normal
distribution in these ways:
1. It is bell-shaped.
2. It is symmetric about the mean.
3. The mean, median, and mode are equal to 0 and are
located at the center of the distribution.
4. The curve never touches the x axis.

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
21
 Characteristics of the t Distribution
The t distribution differs from the standard normal
distribution in the following ways:
1. The variance is greater than 1.
2. The t distribution is actually a family of curves based
on the concept of degrees of freedom, which is
related to sample size.
3. As the sample size increases, the t distribution
approaches the standard normal distribution.

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Degrees of Freedom
• The symbol d.f. will be used for degrees of freedom.
• The degrees of freedom for a confidence interval for the
mean are found by subtracting 1 from the sample size. That
is, d.f. = n – 1.

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
23
x

Confidence Interval for a Population Mean µ, for small samples,

n < 30. Population Standard deviation, σ unknown

If x and s are the mean and standard deviation of a random

sample of size n < 30 from a normal population with mean µ,
a (1-α )100% confidence interval estimate for µ is

x̄ ± t α /2
( ) 𝑠
√𝑛
Where df = n - 1
x
Using t distribution
table

Table 7 contain the rows

that gives the values of t
corresponding to degrees
of freedom, v and the
appropriate tail area α to
the right of the t
distribution.

Example

For a 98% confidence

interval for mean,

α= 0.02
α/2 = 0.01
Sample size, n= 4
Degree of freedom,
v =n-1=3

t0.01,3 = 4.541
 Example : Infant Growth
A random sample of 10 children found that their average growth for the
first year was 9.8 inches. Assume the variable is normally distributed and
the sample standard deviation is 0.96 inch. Find the 95% confidence
interval of the population mean for growth during the first year.

x̄ ± t α /2
9 .8 ± 𝑡 √𝑛
𝑠
0.96
0.05 / 2
( )
√10
0.96
9 .8 ± ( 2.262)
√1 0
9 .8 ± 0.69
9 .11< 𝜇<10.49
Therefore, one can be 95% confident that the population mean of the first-year
growth is between 9.11 and 10.49 inches
Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Example : Home Fires by Candles
The data represent a sample of the number of home fires
started by candles for the past several years. Find the 99%
confidence interval for the mean number of home fires started
by candles each year.
5460 5900 6090 6310 7160 8440 9930

Step 1: Find the mean and standard deviation. The mean is

X = 7041.4 and standard deviation s = 1610.3.
Step 2: Find tα/2 in Table 7. The confidence level is 99%, and the
degrees of freedom d.f. = 6
t 0.005,6 = 3.707.
Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
 Example : Home Fires by Candles
Step 3: Substitute in the formula.

x̄ ± t α /2
7 041.4 ± 𝑡 √𝑛
𝑠
1610.3
0.0 05
( ) √7
1610.3
7 041.4 ±(3.707)
√7
7 041.4 ± 2256.2
4785.2<𝜇< 9297.6
One can be 99% confident that the population mean number of
home fires started by candles each year is between 4785.2 and
9297.6.
Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
Example:

To determine the flow characteristics of oil through a valve, the inlet oil
temperature is measured in degrees Fahrenheit. The following are a sample of 8
readings: 97, 93, 91, 94, 93, 92,89, 90
Construct a 99% confidence interval for the mean inlet oil temperature.

Sample mean, ¯𝑥 =92.375 Sample standard deviation, 𝑠=2.504

α = 0.01 α/2 = 0.005
Sample size, n= 8. v = 8-1=7
t0.005,7 = 3.499

99% confidence interval for the mean inlet oil temperature is

(89.277, 95.473)

We are 99% confident that the mean inlet oil temperature lies between 89.277ºF and
95.473ºF.
 Confidence Intervals for the
Difference Between Two Means/Two Samples

Difference between
means 1   2
of two populations

Independent
Paired samples
samples

Population with Population with

equal variances unequal variances
Independent Samples

Independent samples are samples

mean 1 mean x1
randomly selected from two populations and var iance s12
the selection of one sample from a var iance 12 size n1

population does not affect the selection of

the sample from another population.
Sample 1
Population 1

We are interested in testing the difference in

mean x 2
the mean µ1-µ2 of two normal populations
mean  2 var iance s22
when the variances of both populations are
size n2
unknown. var iance  22

Sample 2
Two different test statistics are used when
the unknown population variances are either Population 2
equal or unequal.
Independent random samples
Confidence interval for the difference between two population means
Variances unknown and equal (Small independent samples)

The (1-α )100% confidence interval estimate for 1   2 is

√
where 2 is the pooled standard error.
2
( 𝑛1 −1 ) 𝑠1 + ( 𝑛 2 − 1 ) 𝑠 2
𝑠𝑝 =
The degree of freedom 𝑛1of+𝑛t 2distribution
−2 is n1 + n2 - 2.

Assumption
 Populations are normally distributed.
 The variances of the two populations, and although unknown, but
they are equal. 2 2
 Two independent random samples of size𝜎 𝜎
1 n1 and n22 ( n1 < 30, n2 < 30).
Exercise

The supervising inspector of incoming quality wants to know if the average

lifetimes of the two brands (brand 1 and brand 2) are the same. Based on past
experience, she believes that the battery lifetimes follow a normal distribution with
equal variance. A sample experiment is conducted: each of ten batteries (five of
each brand) is connected to a test device that places a small drain on the battery
power and records the battery lifetime. The following results (in hours) are
obtained:

Brand 1 43 48 38 41 51
Brand 2 30 26 37 31 34

Construct the 95% confidence interval on the difference between the average
lifetimes of the two brands. Can the supervising inspector of incoming quality
conclude that the average lifetimes of the two brands are equal?
Solution:

√
2 2
( 1
𝑛 −1 ) 1 ( 2
𝑠 + 𝑛 − 1 ) 2
𝑠
𝑠𝑝 =
𝑛1 +𝑛2 −2
= 4.743

( 𝑥¯ 1 − 𝑥¯ 2 ) ± 𝑡 𝛼 𝑠 𝑝
√ 1 1
+
𝑛 1 𝑛2

√
2

1 1
( 44.2− 31.6 ) ± 𝑡0.025 , 8 (4.743) +
5 5
1 2.6 ±(2.306)(4.743) +
1 1
5 5 √
5.682 < 1 2.6 ± 6.918
SPSS Output:
 Notes:
1. If the confidence interval includes 0, we can
say that there is no significant difference
between the means of the two populations, at a
given level of confidence. / 1 - 2 = 0
2. Contain only positive value, there is a
significant difference between the means of the
two populations. / 1 > 2
3. Contain only negative value, there is a
significant difference between the means of the
two populations. / 1 < 2

Copyright © 2015 The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
Confidence interval for the difference between two population
means (Paired Samples)

Definition
Population 1 Population 2

Two samples are said to be

paired or matched
samples when for each x1 y1
data value collected from x2 y2
one sample there is a . .
. .
corresponding data value . .
collected from the second xn yn
sample. Both these data
values are collected from
the same or different
sources. Paired or matched samples
Confidence interval for the difference between two population
means (Paired Samples)
The paired difference 𝑑𝑖 =𝑥𝑖 − 𝑦 𝑖 where xi and yi are values from
sample 1 and sample 2.
Structure of data for a pair comparison.

Pair Sample 1 Sample 2 Difference

1 x1 y1 d1= x1 - y1
2 x2 y2 d2= x2 - y2

n xn yn dn= xn - yn
Confidence interval for the difference between two population
means (Paired Samples)

The (1 – α)100% confidence interval for μd , the mean of population of

differences between the paired samples is

𝑠𝑑
¯ ±𝑡
𝑑 𝛼 /2
√𝑛
where degree of freedom of t is n-1.
The mean and standard deviation of d, the paired differences for the two samples
are calculated using calculator.

Assumption
The distribution of d, difference between paired samples is approximately
normal. .
Exercise

A health spa has advertised a weight-reducing program and has

claimed that the average participant in the program loses is about 17
pounds. A somewhat overweight executive is interested in the
program but is sceptical about the claims and asks for some hard
evidence. The spa allows him to select randomly the records of 10
participants and record their weights before and after the program.
These data are recorded below:
Before 189 202 220 207 194 177 193 202 208 233
After 170 179 203 192 172 161 174 187 186 204

Construct the 99% confidence interval on the mean difference of

weight before and after the program. Does the claim from the health
spa is true? Explain.
Solution
Before 189 202 220 207 194 177 193 202 208 233
After 170 179 203 192 172 161 174 187 186 204
Difference, di 19 23 17 15 22 16 19 15 22 29

𝑑¯ =19.7 𝑠
𝑠𝑑
=4.398
¯ ±𝑡
𝑑
𝑑
𝛼 /2
√𝑛
4.398
1 9.7 ± 𝑡 0.0 05
√ 10
4.398
1 9.7 ± ( 3.25)
√ 10
1 9.7 ± 4.52
15.18< 𝜇<24.22
SPSS Output:

𝑠𝑑
¯ ±𝑡
𝑑 𝛼 /2
√𝑛
4.398
1 9.7 ± 𝑡 0.0 05
√ 10
4.398
𝑠𝑑 1 9.7 ±( 3.25)
√10

¯
𝑑