Scribd
Upload
2 views8 pages

Stat 252 Set 12

This document covers the Student's T-distribution and its applications in statistical analysis, particularly for small sample sizes. It explains the Central Limit Theorem, confidence intervals for the mean, and hypothesis testing, providing examples for clarity. The document also includes specific cases for when the population standard deviation is known or unknown, along with exercises for practice.

Uploaded by

michaelwillis
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 views8 pages

Stat 252 Set 12

This document covers the Student's T-distribution and its applications in statistical analysis, particularly for small sample sizes. It explains the Central Limit Theorem, confidence intervals for the mean, and hypothesis testing, providing examples for clarity. The document also includes specific cases for when the population standard deviation is known or unknown, along with exercises for practice.

Uploaded by

michaelwillis
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/ 8

Set 12: Student’s T-Distribution

Stat 252: February 5, 2025

Recall:Central Limit Theorem (CLT): Let X1 , X2 , . . . , Xn be a random sample of size


n, where n is large (n > 40) from any distribution, with mean µ and standard deviation σ.
Then, √ X −µ
X ∼ Normal(µ, σ/ n) and Z= √ ∼ Normal(0, 1).
σ/ n

Recall: Confidences Intervals for µ


• Case 1: Population σ is known and population is near-normally distributed.

• Case 2: σ is unknown and n is large (n > 40).

• What if σ is unknown and n is small (n ≤ 40)?

Let X1 , X2 , . . . , Xn be a random sample of size n from a (near) normal distribution with mean
µ and unknown variance. Then,

Properties of T Random Variables


• T random variables are continuous (smooth) random variables.
• There are infinitely many T random variables.
▷ Each identified by parameter ν > 0, the degrees of freedom (dof ).
▷ Tν denotes the T -distributed random variable with ν dof.
• The probability density function (pdf) for the T distribution is a symmetric bell curve, always
centred at µ = 0.

• The parameter ν dictates the shape of the T pdf:


▷ As ν increases ⇒

Set 12 1 L.Teshima 2025


Example 1: Determine the following:
(a) For ν = 20, find P (T20 ≥ 1.725).

(b) P (T3 ≤ −4.541).

(c) P (1.943 ≤ T6 ≤ 3.707)

(d) For ν = 15, determine the critical values:

• t15,0.01

Set 12 2 L.Teshima 2025


• t15,0.005

(e) Determine P (T18 > 2.45)

(f) Determine the value c such that P (−c < T10 < c) = 0.90.

Confidence Intervals for the mean µ


• Case 3: σ is unknown and n is small (n ≤ 40).
When the population is near-normal, with unknown standard deviation and a small sample
size, the (1 − α)100% CI is given by:

Note: CIs do exist for when n is small and the population is NOT near-normally distributed - we just don’t
cover them in Stat 252.

Set 12 3 L.Teshima 2025


Example 2: The temperatures from 8 weather stations from around Victoria are measured, and
a sample mean temperature of 14.2◦ C and sample standard deviation of 2.1◦ C are found. Find
the 99% CI for the true mean temperature µ. Assume temperatures are approximately normally
distributed.

Extra Example 1: Seven tomato plants are treated regularly with a standard commercial 5-10-10
fertilizer (5% Nitrogen, 10% Phosphorus, and 10% Potassium). At the end of the growing season,
their total fruit yields are recorded below (in kg). Determine the 90% for the true mean fruit yield
of the fertilized plants.
5.5 4.0 4.1 4.7 5.2 5.8 4.3

Set 12 4 L.Teshima 2025


Hypothesis Tests for µ

For all of the following cases, the hypothesis have the same form:

Right-Tailed: Left-Tailed: Two-Tailed:


H0 : µ = µ0 (or H0 : µ ≤ µ0 ) H0 : µ = µ0 (or H0 : µ ≥ µ0 ) H0 : µ = µ0
H1 : µ > µ0 H1 : µ < µ0 H1 : µ ̸= µ0

▷ Case: Small Sample Size (n ≤ 40)


• Assumptions:

• Test statistic:

• p-values
Right-tailed Test: p-value = P (Tν > tobs )
Left-tailed Test: p-value = P (Tν < tobs )
Two-tailed Test: p-value = 2P (Tν > |tobs |)

Example 3: The giant Pacific octopus (Enteroctopus dofleini ) is noted for being long-lived com-
pared to many other octopus species. Suppose that a certain wild population is known to have
normally distributed lifespans, with an average of 4 years. A researcher wishes to test to the claim
that captive octopuses have a different lifespan. The lifespans (in years) of 6 captive giant Pacific
octopuses are recorded below.
3.4 4.8 5.6 3.9 5.8 5.2
(a) Test the claim that the average lifespan of captive giant Pacific octopus is greater than 4
years.
1. Define the parameter(s) to be tested.

2. Specify H0 and H1 .

3. Specify the test-statistic to be used, and identify its distribution (assuming H0 is true).

Set 12 5 L.Teshima 2025


4. Compute the observed value of the test-statistics.

5. Compute the p-value.

6. Report the strength of evidence against H0 .

(b) Test the claim that the average lifespan of captive giant Pacific octopus is not 4 years.

(c) Create the 90% CI for the population mean lifespan.

Set 12 6 L.Teshima 2025


▷ Case: Normally Distributed Population (with σ known) or Large Sample Size (n > 40)
(See Handbook pages 10-13)

• Assumptions:

• Test statistic:

• p-values
Right-tailed Test: p-value = P (Z > zobs )
Left-tailed Test: p-value = P (Z < zobs )
Two-tailed Test: p-value = 2P (Z < −|zobs |)

Example 4: A grocery chain advertises that customers can get their “Shopping done in 30 minutes
or less”. A survey of 80 shoppers to the chain finds an average trip length of 32 minutes with a
standard deviation of 6 minutes. Is there evidence to suggest that the advertisement is inaccurate?
1. Define the parameter(s) to be tested.

2. Specify H0 and Ha .

3. Specify the test-statistic to be used, and identify its distribution (assuming H0 is true).

Set 12 7 L.Teshima 2025


4. Compute the observed value of the test-statistics.

5. Compute the p-value.

6. Report the strength of evidence against H0 .

Extra Example 2: Repeat the previous example, but with the research question: “Is there evidence
to suggest that the average shopping trip is not 30 minutes?”

Textbook Readings: Subsection 8-4a on the Student t distribution, Intro to Ch 12, and 12-1.
Problems (do by next class): 8.111, 8.112, 12.4, 12.8, 12.10, 12.12, 12.16, 12.20, 12.24, 12.26, 12.30

Set 12 8 L.Teshima 2025

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy