|
Activity Number:
|
369
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #301338 |
|
Title:
|
Determining Minimum Sample Sizes To Achieve Central Limit Theorem Closeness When Sampling from Various Populations
|
|
Author(s):
|
Jamis Perrett+ and Steve Hoff*+
|
|
Companies:
|
Texas A&M University and University of Northern Colorado
|
|
Address:
|
3143 TAMU, College Station, TX, 77843-3143, CB 124, Greeley, CO, 80639,
|
|
Keywords:
|
Normality Assumption ; Minimum Sample Size ; Central Limit Theorem
|
|
Abstract:
|
One version of the Central Limit Theorem states that when repeatedly sampling independently from any distribution with finite first and second moments, and for some large sample size (n), the sample mean possesses an approximate Gaussian distribution. How close is "approximate" and how large is "large"? Using techniques from mathematical statistics and from simulation sampling, the researchers have developed a technique to assess minimum sample sizes which are sufficient to insure approximate closeness for selected populations both symmetric and skewed, with domains both finite and infinite, and with both light and heavy tails. Simulation routines are designed in a manner consistent with minimum sample sizes determined from theoretical results.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2008 program |
