Alternatives to Cut-Off Sampling for a Family of Surveys

• Activity Number: 243
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Government Statistics
• Author(s): Paula Ellen Mason*+ and Benita Jean O'Colmain and Pedro Saavedra and Amerine Woodyard and Bin Zhang
• Companies: Energy Information Administration and ICF Macro and ICF Macro and Energy Information Administration and Energy Information Administration
• Keywords: benchmarking to lagged values ; sampling from skewed distributions ; ratio scale ; Chromy allocation algorithm

Abstract

The respondents to monthly energy surveys are used as the sampling frames for corresponding weekly supply surveys. The weekly surveys are based on a cut-off sampling approach that account for 90% of the population volumes of the monthly surveys. A ratio estimator which uses the monthly data is used to estimate the weekly population volumes. Recent changes in the surveys and reporting level have resulted in much larger sample sizes. This paper describes some alternative sampling approaches and the analysis performed in the development of new sampling and estimation methodology for the weekly surveys. The analysis is mainly focused on precision of publication estimates which span individual surveys, and sample size. The precision of the publication estimates is measured with bias, mean absolute deviation, and mean square deviation.