|
Activity Number:
|
31
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Government Statistics
|
| Abstract - #306083 |
|
Title:
|
The Use of Geocoding to Locate Outlets Outside of Sample Area Boundaries to Determine Significant Areas of Commerce
|
|
Author(s):
|
John Schilp*+ and Fred Marsh, III
|
|
Companies:
|
Bureau of Labor Statistics and Bureau of Labor Statistics
|
|
Address:
|
2 Massachusettes Ave., NE, Washington, DC, 20212,
|
|
Keywords:
|
geocoding ; sampling ; index calculation
|
|
Abstract:
|
This paper gives an introduction to the method of sampling retail outlets in the U.S. Consumer Price Index (CPI). Some of these sampled retail outlets are located outside the primary sampling area boundary and are currently priced. To reduce data collection costs, BLS would like to develop a rule to eliminate some outlets from being priced. Currently, BLS prices any selected outlet that is less than 25 miles from the primary sample area boundary or has more than 10 quotes in a cluster. We examine the impact of eliminating certain groups of outlets from the sample on the CPI index and also explain the definition of a spatial cluster and the new "significant area of commerce."
|
