Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 210 - Contributed Poster Presentations: Survey Research Methods Section
Type: Contributed
Date/Time: Tuesday, August 4, 2020 : 10:00 AM to 2:00 PM
Sponsor: Survey Research Methods Section
Abstract #314096
Title: Benchmarking Nonsampled Companies’ Imputed Values of Net Metering Capacity for the Monthly Electric Power Industry Report
Author(s): David Kinyon*
Companies: Energy Information Administration
Keywords: establishment surveys; imputation; time series methods; benchmarking; modified Denton method
Abstract:

The U.S. Energy Information Administration (EIA) releases the Monthly Electric Power industry Report to publish monthly estimates of retail sales and revenue, net generation, and capacity by end-use sector for the electric power industry. For this report, data are collected from a sample of utility and nonutility companies, and data for nonsampled companies are imputed using separate regression models by end-use sector. Annual data for both sampled and nonsampled companies are collected for the Annual Electric Power Industry Report that is published by EIA, and the imputed monthly values of net metering capacity for December are typically somewhat different from the corresponding annual reported values. EIA researched different methods of adjusting individual nonsampled companies’ imputed monthly values of net metering capacity for 2017 before these values were input to the monthly imputation procedure for 2018. This paper will discuss this research including EIA’s decision to use the modified Denton method to adjust these monthly imputed values to the reported annual benchmarks while maintaining, as closely as possible, the month-to-month changes in the imputed series.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program