Non-Gaussian uncertainty distributions: historical trends and forecasts of the United States energy sector, 1983-2010

摘要

Applies a novel method of uncertainty parameterization and analysis to time-series data of recent supply and demand projections for the United States energy sector. The characterization of uncertainty, particularly for low probability events and application of the method to an existing set of forecasts are considered. Based on determinations of the actual uncertainties in past forecasts of over 170 energy producing and consuming sectors of the US economy, a simple one-parameter model is presented that can be used to estimate a probability distribution for future projections.