Scribd
Upload
12 views11 pages

Chapter9 - Inf - Varaince - Student

Chapter 9 discusses hypothesis testing for two population variances, focusing on the importance of variances in statistical inference. It outlines the null and alternative hypotheses, the use of the F-test statistic, and the conditions required for its application. Additionally, it provides R code for calculating p-values and critical values for hypothesis testing.

Uploaded by

minulo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
12 views11 pages

Chapter9 - Inf - Varaince - Student

Chapter 9 discusses hypothesis testing for two population variances, focusing on the importance of variances in statistical inference. It outlines the null and alternative hypotheses, the use of the F-test statistic, and the conditions required for its application. Additionally, it provides R code for calculating p-values and critical values for hypothesis testing.

Uploaded by

minulo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Chapter 9

Inference for Variances

STAT 190
Introduction to Statistical Methodology

Dr. Aruni Jayathilaka


Fall 2024

1
Hypothesis Tests for Two Population Variances

• Primarily, inference tends to be focused on means (and proportions, which we will learn later).
• Variances, though, are essential to inference of means.
• We may be interested in determining whether two population variances are equal.
• Consider the case of wanting to compare two populations.
• We have talked about testing whether two means of these independent populations are equal to each
other.
• We also can test whether the variances of the two populations are equal to each other.

2
Hypothesis Tests for Two Population Variances
• Null hypothesis: population variances are equal.
• We attempt to see if there is a difference in variances.

• Two-tailed test:
H0: 𝜎12 = 𝜎22 vs H1: 𝜎12 ≠ 𝜎22

• Right-tailed test: (upper tail)


H0: 𝜎12 ≤ 𝜎22 vs H1: 𝜎12 > 𝜎22

• Left-tailed test: (lower tail)


H0: 𝜎12 ≥ 𝜎22 vs H1: 𝜎12 < 𝜎22

3
F distribution

• The F-distribution is a skewed distribution.

• Its support be the set of positive (non-negative) real


numbers.

• As the shape of an F-distribution depends on two sets of


degrees of freedom as parameters 𝑑𝑓1 and 𝑑𝑓2 .
• Increase in the degrees of freedom 𝑑𝑓1 and 𝑑𝑓2 , its
skewness decreases.

4
Hypothesis Tests for Two Population Variances
• To test such hypotheses, we use an F test statistic.
• In order to perform a F test of two variances, it is important that the following are true:
• The populations from which the two samples are drawn are approximately normally distributed.
• The two populations are independent of each other.

• Take 𝜎12 and 𝜎22 be the unknown population variances and 𝑠12 and 𝑠22 be the sample variances. Let
the sample sizes be 𝑛1 and 𝑛2 .
• We are interested in comparing the two sample variances, we use the F ratio:

𝑠12 /𝜎12
𝐹𝑜𝑏𝑠 = 2 2
𝑠2 /𝜎2

• This F statistic follows F distribution with (𝑛1 – 1) is the degrees of freedom for the numerator and
(𝑛2 – 1) is the degrees of freedom for the denominator.
5
Hypothesis Tests for Two Population Variances

• Under the null hypothesis 𝝈𝟐𝟏 = 𝝈𝟐𝟐 , so we have the following:

𝑠12
𝐹𝑜𝑏𝑠 = 2 ~𝐹𝑛1−1,𝑛2−1
𝑠2

• Once we have our test statistic, we can compare to the F distribution to calculate a p-value and critical
value.

6
Hypothesis Tests for Two Population Variances
R codes: P-values
• One-tailed, lower (left) tail:
𝑝 = 𝑃 𝐹 < 𝐹𝑜𝑏𝑠
pf(F_obs, n1-1, n2-1)

• One-tailed, upper (right) tail;


𝑝 = 𝑃 𝐹 > 𝐹𝑜𝑏𝑠
1-pf(F_obs, n1-1, n2-1)

• Two-tailed :
if 𝐹𝑜𝑏𝑠 ≤ 1; 2*pf(F_obs, n1-1, n2-1)
if 𝐹𝑜𝑏𝑠 > 1 ; 2* (1-pf(F_obs, n1-1, n2-1))

7
Hypothesis Tests for Two Population Variances

• If the p−𝐯𝐚𝐥𝐮𝐞 ≤ 𝜶 , reject 𝑯𝟎 and conclude that the population variances are unequal to each other.

• If the p−𝐯𝐚𝐥𝐮𝐞 > 𝜶 , fail to reject 𝑯𝟎 and conclude that that population variances are equal to each
other.

8
Hypothesis Tests for Two Population Variances
Also, Instead of p-values, we can use critical values for rejection.
R codes: Critical Values
• One-tailed, lower (left) tail:
Reject if 𝐹 ∗ ≤ 𝐹𝑛1−1,𝑛2−1,𝛼
qf(alpha, n1-1, n2-1)

• One-tailed, upper (right) tail:


Reject if 𝐹 ∗ ≥ 𝐹𝑛1−1,𝑛2−1,1−𝛼
qf(1-alpha, n1-1, n2-1)

• Two-tailed:
Reject if 𝐹 ∗ ≤ 𝐹𝑛1−1,𝑛2−1,𝛼/2 or Reject if 𝐹 ∗ ≥ 𝐹𝑛1−1,𝑛2−1,1−𝛼/2
qf(alpha/2, n1-1, n2-1) or qf(1-alpha/2, n1-1, n2-1) 9
Hypothesis Tests for Two Population Variances
Example:
We are interested in exploring daily caloric intake for two populations: college professors and college
students. In particular, we want to explore the hypothesis,

Take a sample of 𝑛1 = 10 professors and 𝑛2 = 20 students.


The sample variances are 𝑠12 = 2.33 and 𝑠22 = 9.97.

Calculate the test statistic:

Since 𝐹𝑜𝑏𝑠 ≤ 1;

10
Hypothesis Tests for Two Population Variances
Example:

11

You might also like