Efficiently Utilizing Available Regressor Data Through a Multi-Tiered Survey Estimation Strategy

by Joel Robert Douglas.

Abstract: Efficiently utilizing available regressor data is an important goal of an effective survey sampling campaign, and choosing amongst competing regression models is an important part of the survey statistician’s job. The multi-tiered estimation framework initially developed for use on U.S. Energy Information Administration Form EIA-923, “Power Plant Operations Report” provides a logical organization of available estimation choices, and codifies their order of importance. One key concept associated with a multi-tiered estimation strategy is that regression options should not be ignored simply because they are not available universally to all members of the population. This concept expands the set of available regression models, which can lead to publishing more accurately estimated data totals.

Key Words: Establishment survey, quasi-cutoff sampling, regression, prediction, relative standard error,

Joel Robert Douglas, douglas.joel@gmail.com

Editor: James R. Knaub, james.knaub@eia.gov READING THE ARTICLE: You can read the article in portable document (.pdf) format (508666 bytes.)
If you have any comments or for further discussion, contact an author.

NOTE: This paper was revised on October 10, 2013.

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 1087 times since SEPTEMBER 25 2013.

Return to the InterStat Home Page.