Title :
Non-Gaussian uncertainty distributions: historical trends and forecasts of the United States energy sector, 1983-2010
Author :
Kammen, Daniel M. ; Shlyakhter, Alexander I. ; Broido, C.L. ; Wilson, Richard
Author_Institution :
Dept. of Phys., Harvard Univ., Cambridge, MA, USA
Abstract :
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
Keywords :
economic cybernetics; energy resources; forecasting theory; history; probability; time series; uncertain systems; US economy; United States energy sector; forecasts; future projections; historical trends; low probability events; nonGaussian uncertainty distributions; one-parameter model; probability distribution; supply and demand projections; time-series data; uncertainty parameterization; Carbon dioxide; Data analysis; Demand forecasting; Economic forecasting; Load forecasting; Physics; Predictive models; Probability distribution; Supply and demand; Uncertainty;
Conference_Titel :
Uncertainty Modeling and Analysis, 1993. Proceedings., Second International Symposium on
Conference_Location :
College Park, MD
Print_ISBN :
0-8186-3850-8
DOI :
10.1109/ISUMA.1993.366782
