Estimation for Detailed Publication Levels in the Current Employment Statistics Survey

• Activity Number: 414
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Julie Gershunskaya*+
• Companies: Bureau of Labor Statistics
• Keywords: small area estimation ; area-level model

Abstract

One prerequisite for successful small domain modeling is the ability to form a "pool" of like "areas", such that "borrowing strength" across the areas is a viable concept. This may not always be possible for various reasons. We consider estimation of employment from the Current Employment Statistics (CES) survey conducted by the U.S. Bureau of Labor Statistics. Detailed estimation cells are defined as intersections of industrial and geographic levels. Combining direct detailed level estimates across States is not always feasible due to possible substantial differences between States, intricacies of individual States' estimation structures, or simply due to logistics of the production procedures at State levels (for example, differences in the production timeframe.) A simple area-level model is formulated for higher level estimates. We explore the possibility of applying the parameters from the higher-level setting to States' detailed levels.