|
Activity Number:
|
603
- The Importance of Statistical Analysis in Energy Modeling
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 3, 2017 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Government Statistics Section
|
|
Abstract #323130
|
View Presentation
|
|
Title:
|
Bayesian Dynamic Linear Models for Spot Prices of Hydrocarbon Gas Liquids
|
|
Author(s):
|
Peter Gross*
|
|
Companies:
|
U.S. Energy Information Administration
|
|
Keywords:
|
Bayesian ;
dynamic linear model ;
propane ;
hydrocarbon gas liquids ;
prices ;
ethane
|
|
Abstract:
|
Hydrocarbon gas liquids (HGLs) include a range of substances produced during natural gas processing and petroleum refining. Advances in horizontal drilling and hydraulic fracturing ("fracking") technologies have recently led to record high levels of HGL production in the United States. The U.S. Energy Information Administration (EIA) projects HGL prices based on long-term supply and demand effects estimated from historical data. These effects are now changing due to the technological advances and subsequent market reactions. We present a multi-process Bayesian Dynamic Linear Model (DLM) developed to forecast certain HGL prices based on a combination of expert judgment and historical data. We discuss results of a simulation study designed to test the model's performance under several energy market scenarios.
|
Authors who are presenting talks have a * after their name.
