Scribd
Upload
2 views12 pages

Lecture 5c. Inference About The Population Variance

The document discusses the inference about population variance, focusing on the use of the Chi-Square distribution for constructing confidence intervals. It provides formulas for estimating confidence intervals for variance and includes examples of calculating these intervals based on sample data. Additionally, it covers hypothesis testing for population variance using Chi-Square statistics.

Uploaded by

munesubishi13
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
2 views12 pages

Lecture 5c. Inference About The Population Variance

The document discusses the inference about population variance, focusing on the use of the Chi-Square distribution for constructing confidence intervals. It provides formulas for estimating confidence intervals for variance and includes examples of calculating these intervals based on sample data. Additionally, it covers hypothesis testing for population variance using Chi-Square statistics.

Uploaded by

munesubishi13
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/ 12

Inference About A Population

Variance

G3512ES/EST3672 5.3.1
Introduction
• The last parameter we need to make inference
about is the population variance (σ2).
• There are many instances where we might be
interested in knowing something about the spread
of a population based on a sample.
• For example, we might believe that a particular
group of students seems to have a wider variation in
their grades than those from the past.
• Or a part manufacturer may be concerned that one
of the parts it's manufacturing is too inconsistent,
even though the mean may be at specifications.
G3512ES/EST3672 5.3.2
The Chi-Square (𝜒 2 ) distribution
• Before we can develop a confidence interval
for the variance, we need another
distribution, the 𝜒 2 distribution.
• If a random sample size n is obtained from a
normally distributed population with mean μ
and standard deviation σ, then
2 (𝑛−1)𝑆 2
• 𝜒 = has a chi-square distribution
𝜎2
with n-1 degrees of freedom.

G3512ES/EST3672 5.3.3
PROPERTIES OF CHI-SQUARE DISTRIBUTION

• It is not symmetric
• The shape depends on the degrees of freedom.
• As the number of degrees of freedom increases, the distribution becomes
more symmetric.
• Χ2≥0

G3512ES/EST3672 5.3.4
2
Confidence Interval Estimator for 𝜎
• The formula for constructing confidence
interval.

• 𝜒 2 (1−𝛼) < 𝜒 2 < 𝜒 2 𝛼


2 2

(𝑛−1)𝑆 2
• Substitute 𝜒 2 = into the formula
𝜎2
above and manipulate to get the confidence
interval estimator for 𝜎 2
G3512ES/EST3672 5.3.5
2
Confidence Interval Estimator for 𝜎
(𝑛−1)𝑆 2 (𝑛−1)𝑆 2
• < 𝜎2 <
𝜒2 𝛼/2 𝜒2 1−𝛼/2

(𝑛−1)𝑆 2
• = Lower Confidence Limit (LCL)
𝜒2 𝛼/2

(𝑛−1)𝑆 2
• = Upper Confidence Limit (UCL)
𝜒2 1−𝛼/2

G3512ES/EST3672 5.3.6
Example 1
• Suppose a sample of 30 students are given an
IQ test. If the sample has a standard deviation
of 12.23 points, find a 90% confidence interval
for the population variance.
(𝑛−1)𝑆 2 30−1 ∗12.232 4337.6141
• LCL = = =
𝜒2 𝛼/2 𝜒2 0.05 42.557

= 101.9248

G3512ES/EST3672 5.3.7
Example 1
• Suppose a sample of 30 students are given an
IQ test. If the sample has a standard deviation
of 12.23 points, find a 90% confidence interval
for the population variance.
(𝑛−1)𝑆 2 30−1 ∗12.232 4337.6141
• UCL = = =
𝜒2 1−𝛼/2 𝜒2 0.95 17.708

= 244.9522

G3512ES/EST3672 5.3.8
2
Example 2: Hypothesis testing for 𝜎
• In a sample of 25 cereal boxes, the sample
standard deviation is 17.7 grams. Supposing
that the sample is drawn from a normal
distribution, test, at 5% significance level
whether the population variance differs from
225.

G3512ES/EST3672 5.3.9
2
Example 2: Hypothesis testing for 𝜎
•𝐻0 : 𝜎 2 = 225
•𝐻1 : 𝜎 2 ≠ 225

• Chi-squared statistic
25−1 17.72
• 𝜒2 = = 33.42
225
• Chi-squared critical value
•𝜒 2 1−0.05 ,24 = 12.40 and 𝜒 2 0.05
,24
= 39.36.
2 2

G3512ES/EST3672 5.3.10
2
Example 2: Hypothesis testing for 𝜎

• Reject H0 if chi-square statistic is less than


χ21−α/2 or reject H0 if chi-square statistic is
greater than χ2 α/2
• Conclusion
• χ2 stat = 33.42 > χ21−α/2 (12.4), do not reject
H0
• χ2 stat = 33.42 < χ2 α/2 (39.6), do not reject H0

G3512ES/EST3672 5.3.11
Ex.1
a) A sample of 51 elements is selected to estimate a
95% confidence interval for the variance of the
population. Find the chi-square values to be used
for this interval estimation.
b) A sample of 20 cans of tomato juice showed a
standard deviation of 0.4 ounces. A 95%
confidence interval estimate of the variance for
the population will be?

G3512ES/EST3672 5.3.12

You might also like