9'3: Central Limit Theorem - PowerPoint PPT Presentation

1 / 11
About This Presentation
Title:

9'3: Central Limit Theorem

Description:

Put your 25 pennies in a cup. Randomly select 5 pennies. ... Now, we will look at a histogram of all the distributions. You are going to come up and put a dot ... – PowerPoint PPT presentation

Number of Views:29
Avg rating:3.0/5.0
Slides: 12
Provided by: peterj72
Category:
Tags: central | cup | limit | theorem

less

Transcript and Presenter's Notes

Title: 9'3: Central Limit Theorem


1
9.3 Central Limit Theorem
  • Many populations do indeed have an approximately
    normal distribution. But, what about
    distributions that are skewed or not normal???
    If the sampling distribution of x-bar is not
    approximately normal, we have no way of using
    x-bar to compare with the population parameter.

2
  • Look at your 25 pennies. If we were to plot the
    age of your 25 pennies, what kind of distribution
    do you think it would resemble? Symmetric or
    skewed?
  • Fill out the table to find the distribution of
    your coins. Remember we are using age as the
    x-variable.

3
  • Put your 25 pennies in a cup. Randomly select 5
    pennies. Find the average age of your sample of
    five. Record this as x-bar(5). Repeat this to
    get another random sample of size 5.
  • Now, randomly select 10 pennies. Find the
    average age and record this as x-bar(10). Repeat
    this process to get another random sample of size
    10.

4
  • Find the average age of all 25 of your coins.
    Record this as x-bar(25).
  • Now, we will look at a histogram of all the
    distributions. You are going to come up and put
    a dot to represent each pennys age in your
    population.

5
  • Population Distribution

6
  • Sampling Distribution for n 5

7
  • Sampling Distribution n 10

8
  • Sampling Distribution n 25

9
  • Central Limit Theorem Simulation

10
  • Central Limit Theorem Draw an SRS of size n
    from any population with mean µ and finite std.
    dev. s. When n is large enough the sampling
    distribution of the sample mean, x-bar, is close
    to the normal distribution.
  • How large a sample is needed for x-bar to be
    close to normal depends on the population
    distribution. More observations are required for
    distributions farther from normal.

11
  • Ex. The time a technician requires to perform
    preventative maintenance on an A/C unit is
    described by the exponential distribution with µ
    1 and s 1. You have 70 units in your office
    building, what is the probability that their
    average maintenance time is more than 50 minutes?
Write a Comment
User Comments (0)
About PowerShow.com