|
Activity Number:
|
268
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics and Marketing
|
| Abstract - #301971 |
|
Title:
|
Statistical Assessment of the Sample Size Requirement for Projection of the Automobile Sales Volume in the U.S. Market
|
|
Author(s):
|
Keiko I. Powers*+
|
|
Companies:
|
J.D. Power and Associates
|
|
Address:
|
324 Hickory Grove Drive, Thousand Oaks, CA, 91320,
|
|
Keywords:
|
sample size requirements ; confidence intervals ; negative binomial distribution ; sales volume projection ; US automobile market
|
|
Abstract:
|
The paper illustrates how a confidence interval approach based on the negative binomial distribution is used for determining the sample size requirement for projecting the automobile sales in the U.S. market. For the sales projection, it is important to know the relationship between the number of automobile dealers in the sample and the accuracy of the estimated sales volume. The sales volume trend is very volatile, and therefore it is clear that a relatively large number of dealers' data are needed for accurate sales projection. However, because the overall sales volume varies considerably among dealers, statistically deriving sample size requirements called for a new methodology. The paper provides a step-by-step description of the approach and demonstrates its performance by using actual vehicle point-of-sales data collected from several thousand dealers in the U.S. market.
|
