Scribd
Upload
71 views

Chapter7 Stats

The document summarizes key aspects of the central limit theorem (CLT). It discusses how the CLT states that the distribution of sample means approaches the normal distribution as sample size increases, regardless of the underlying population distribution. The CLT allows us to determine probabilities and percentiles involving sample means. An example is provided to illustrate how to use the CLT to test if a sample mean provides statistically significant evidence that the population mean has changed over time.

Uploaded by

Poonam Naidu
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
71 views

Chapter7 Stats

The document summarizes key aspects of the central limit theorem (CLT). It discusses how the CLT states that the distribution of sample means approaches the normal distribution as sample size increases, regardless of the underlying population distribution. The CLT allows us to determine probabilities and percentiles involving sample means. An example is provided to illustrate how to use the CLT to test if a sample mean provides statistically significant evidence that the population mean has changed over time.

Uploaded by

Poonam Naidu
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/ 3

CHAPTER 7: THE CENTRAL LIMIT THEOREM

Lecture Notes for Introductory Statistics1

Daphne Skipper, Augusta University (2016)

1. The Central Limit Theorem For Means


1.1. CLT Setup. Suppose we know that in 2005, the average number of points
scored in an NFL game (combined score of both teams) was 42 with a standard
deviation of 8 points. We wonder if the scoring has changed over the last decade,
so we take a random sample of thirty games in the 2015 season and calculate a
sample mean of 46.8 points per game.
Of course this makes us suspect that the average number of points per game
has increased, but we only have partial 2015 data, so it is not enough simply to
compare the means. We rephrase the question:

“Is there statistical evidence that there are more points scored per game now
than there were in 2005?”

To answer this question, we have to decide if the sample mean of 46.8 is very
different from what we would expect if the (population) mean number of points
is still 42, like in 2005. In other words, we need to see if P (x ≥ 45) is very
small, assuming the population mean number of points per game is still 42. But to
calculate this probability, we need the distribution of X, rather than the distribution
of X! This is where the Central Limit Theorem comes in.

In this example, we have the random variables X and X:

X = the number of points scored in an NFL game in 2015,


X = the mean (average) number of points scored in a random sample of n = 30
NFL games in 2015.
The Central Limit Theorem says that if X has mean µ = 42 points (our assumed
average) andstandard
 deviation σ = 8 points (our assumed standard deviation),
then X ∼ N 42, √830 .

1.2. CLT Statement. Under certain conditions, the Central Limit Theorem pro-
vides the distribution of the random variable defined to be the means of samples of
a certain size from a distribution whose mean and standard deviation are known.

The Central Limit Theorem says that if

X = a random variable (any distribution) with mean µX and standard deviation


σX , and
X = the mean of a random sample of n data values from X,

1
These lecture notes are intended to be used with the open source textbook “Introductory
Statistics” by Barbara Illowsky and Susan Dean (OpenStax College, 2013).
1
Chapter 7 Notes The Central Limit Theorem D. Skipper, p 2

 
σX
then the distribution of X approaches the distribution N µX , √ n
as n gets large.

Let’s take a closer look at the requirements for the Central Limit Theorem. The
following are true:
σX
• Regardless of n, it is always true that µX = µX and σX = √ n
.
• Regardless n, if X is normally distributed, then X is also normally dis-
tributed.
• If X is not normally distributed, the distribution of X approaches a normal
distribution as n gets large.
• How big is “large”? Many textbooks recommend that if X is normally
distributed OR n ≥ 30, then we can assume X is normally distributed .

1.3. CLT Intuition. The Central Limit Theorem is actually quite intuitive. For
larger and larger sample sizes, we would expect (by the Law of Large Numbers) for
the sample mean x to get closer and closer to the population mean µX . So it makes
sense that the mean of X is the same as the mean of X, and the bigger the sample
size, the smaller the standard deviation (or standard error) of X. In other words,
the sample mean is a better and better approximator of the population mean as
the sample size increases.

2. The Central Limit Theorem For Sums


We aren’t doing this section.

3. Applying the Central Limit Theorem (For Means)


In this section we apply the Central Limit Theorem the find probabilities and
percentiles involving X.
Example 1. Finding probabilities involving X and X. The length of time
taken on the SAT for a group of students is normally distributed with a mean of
2.5 hours and a standard deviation of 0.25 hours. A sample size of n = 60 is drawn
randomly from the population.
(1) X =
(2) X ∼
(3) X =
(4) X ∼
(5) Find the probability that a single student takes between 2 and 3 hours.
(6) Find the probability that the sample mean of 60 students is between 2 and
3 hours.
TryIt 7.2.

Example 2. Probabilities and Percentiles involving x. In a study reported


Oct. 29, 2012 on the Flurry Blog, the mean age of tablet users was 34 years.
Suppose the standard deviation is 15 years. Take a sample of size n = 100.
(1) What are the mean and standard deviation for the sample mean ages of
tablet users?
(2) What does the distribution look like?
(3) Find the probability that the sample mean age is 30 or older.
Chapter 7 Notes The Central Limit Theorem D. Skipper, p 3

(4) Find the 10th percentile for the sample mean age.
(5) Find the 95th percentile for the sample mean age.
Example 7.3.

Let’s return to the example from the beginning of the notes for this chapter...
Example 3. Statistically significant result? Suppose we know that in
2005 the average number of points scored in an NFL game (combined score of both
teams) was 42 with a standard deviation of 8 points. We wonder if the scoring has
changed over the last decade, so we take a random sample of 30 games in the 2015
season and calculate a sample mean of 46.8 points per game. Let X = the number
of points scored in a 2015 NFL game.
(1) We will assume that X has a mean and standard deviation equal to those
from the 2005 NFL season. (Why are we making this assumption about
X?)
(2) Based on our assumption about X, X ∼
(3) Based on our assumption about X, P (X ≥ 46.8) =
(4) Describe the significance of the probability from part (2).
(5) Is there statistical evidence that there are more points scored now than in
2005? Explain.

4. Stats Lab
This is a good time for a project to see the Central Limit Theorem in action.
The Cookie Recipe Stats Lab at the end of Chapter 7 is a good option.

You might also like