Short-Term Energy Outlook Model Documentation: Natural Gas Consumption and Prices
June 25, 2015
Overview
The natural gas consumption and price modules of the Short-Term Energy Outlook (STEO) model are designed to provide consumption and end-use retail price forecasts for the residential, commercial, and industrial sectors in the nine Census districts (Figure 1) and natural gas working inventories in three regions (Figure 2). Natural gas consumption shares and prices in each Census district are used to calculate an average U.S. retail price for each end-use sector.
In 2010, EIA implemented revisions to the EIA-857 natural gas survey to collect data on monthly sendout volumes by local distribution companies (LDCs) and pipelines in addition to billing volumes. Sendout data, which reflects actual deliveries rather than billed deliveries, provides a more accurate measure of aggregate natural gas deliveries during each calendar month. Sendout-based data were first reported in the Natural Gas Monthly in the October 2010 edition, which reports data through August 2010.
STEO implemented several changes because of the survey changes. Prior to the changes, U.S. residential, commercial, and industrial sector consumption forecasts were derived by summing regional consumption forecasts. Currently, the regional consumption volume forecasts are used only as weights in determining U.S. average national residential, commercial, and industrial prices. Once a sufficient number of observations of consistent (sendout-basis) history are available, the STEO model may be revised to calculate U.S. consumption volumes by summing regional consumption. The frequency of the STEO model is monthly and the model equations are used to produce monthly forecasts over a 13-to-24 month horizon (every January the STEO forecast is extended through December of the following year).
The STEO model contains over 2,000 equations, of which about 450 are estimated regression equations. The regression equations are estimated and the forecast models are solved using the IHS EViews Econometric Software. The natural gas consumption and price modules contain 145 equations, of which 61 are estimated regression equations. Some input variables to the STEO model are exogenous, coming from other modules in the STEO model or forecasts produced by other organizations (e.g., weather forecasts from the National Oceanic and Atmospheric Administration).