Understanding errors in EIA projections of energy demand

Fischer, Carolyn; Herrnstadt, Evan; Morgenstern, Richard D.

doi:10.1016/j.reseneeco.2009.04.003

Cited by 26 publications

(14 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, ill-defined forecast objectives are also bad for consumers of forecasts such as policy makers. This is a point made by others; Donkor et al (2012) points to Auffhammer (2007) and Fischer et al (2009) for their work analyzing the performance of EIA energy forecasts.…”

Section: Which Forecasting Methods Are Currently Used By Urban Water mentioning

confidence: 89%

Forecasting Residential Water Consumption in California: Rethinking Model Selection

Buck

Auffhammer

Soldati

et al. 2020

Water Resources Research

View full text Add to dashboard Cite

Urban water managers use forecasts of water consumption to determine management decisions and investment choices. Public reports show that water utilities rely on forecast models that are not selected based on their out-of-sample prediction performance; further, these reports frequently only present a single forecast instead of a range of forecasts. In our review of the academic literature on forecasting long-term water consumption, only a few analyses consider out-of-sample prediction performance measures to assess prediction ability. In none of these analyses did out-of-sample prediction performance drive model selection. Ensemble-type long-term forecasts based on multiple models were also lacking. Using annual data on single-family residential water consumption in Southern California, we show that predictive ability varies drastically depending on how the forecast model is selected. As an illustration of how forecast performance is affected when the criteria for model selection does support forecasting objectives, we compare statistical models with the best in-sample and out-of-sample prediction performance. We find that the models with the best in-sample performance over-estimate consumption five years out by 10%-25% compared to actual consumption. In contrast, the top 1% of models selected based on out-of-sample prediction performance came within 1% of actual consumption. Finally, we compare the performance of our ensemble-type forecasts to those reported in public documents derived from models selected based on non-out-of-sample prediction performance criteria. Our results highlight the benefits of (i) using an out-of-sample evaluation criterion to guide model selection and (ii) reporting ensemble forecasts in lieu of a single forecast. Key Points:• Our forecasting method makes out-of-sample prediction performance the criterion for model selection • Models selected on out-of-sample performance drastically outperform forecast models employed in practice • We predict residential water consumption in 2035 for Southern California, consumption decreases 0.5 percent to over a 10 percent reduction

show abstract

Section: Which Forecasting Methods Are Currently Used By Urban Water mentioning

confidence: 89%

Forecasting Residential Water Consumption in California: Rethinking Model Selection

Buck

Auffhammer

Soldati

et al. 2020

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…For example, the USFS projections assume that water‐use efficiency improvements will continue in the municipal, industrial, and thermoelectric sectors throughout the projection period, and the EIA assumes that energy crop production will be limited to areas that require no irrigation. A recent study found a “fairly persistent tendency” by EIA over the past 22 years to underestimate total energy demand by an average of 2% per year over the short term, and that projection errors for individual fuel production and energy consuming sectors can be considerably larger (up to 7%), and can vary by time period, reflecting the idea that projection errors can increase with projection horizon (Fischer et al. , 2009).…”

Section: Resultsmentioning

confidence: 99%

“…sectors throughout the projection period, and the EIA assumes that energy crop production will be limited to areas that require no irrigation. A recent study found a ''fairly persistent tendency'' by EIA over the past 22 years to underestimate total energy demand by an average of 2% per year over the short term, and that projection errors for individual fuel production and energy consuming sectors can be considerably larger (up to 7%), and can vary by time period, reflecting the idea that projection errors can increase with projection horizon (Fischer et al, 2009). The study suggested that EIA generally underpredicted demand for residential and commercial electricity, natural gas for electricity, and gasoline for transportation; and overpredicted demand for other natural gas uses, industrial electricity, coal, and renewable fuels.…”

Section: Caveatsmentioning

confidence: 99%

Future U.S. Water Consumption: The Role of Energy Production¹

Elcock

2010

J American Water Resour Assoc

View full text Add to dashboard Cite

Elcock, Deborah, 2010. Future U.S. Water Consumption: The Role of Energy Production. Journal of the American Water Resources Association (JAWRA) 46(3):447‐460. DOI: 10.1111/j.1752‐1688.2009.00413.x Abstract: This study investigates how meeting domestic energy production targets for both fossil and renewable fuels may affect future water demand. It combines projections of energy production developed by the U.S. Department of Energy with estimates of water consumption on a per‐unit basis (water‐consumption coefficients) for coal, oil, gas, and biofuels production, to estimate and compare the domestic freshwater consumed. Although total domestic freshwater consumption is expected to increase by nearly 7% between 2005 and 2030, water consumed for energy production is expected to increase by nearly 70%, and water consumed for biofuels (biodiesel and ethanol) production is expected to increase by almost 250%. By 2030, water consumed in the production of biofuels is projected to account for nearly half of the total amount of water consumed in the production of all energy fuels. Most of this is for irrigation, and the West North Central Region is projected to consume most of this water in 2030. These findings identify an important potential future conflict between renewable energy production and water availability that warrants further investigation and action to ensure that future domestic energy demand can be met in an economically efficient and environmentally sustainable manner.

show abstract

“…In order to appreciate the surprise and shock caused by the high oil prices during 2008 and by today's levels of above $100/b, it is worth recalling past expert assessments, e.g. : US Department of Energy (2001) and International Energy Agency (2004) were both promising a doubling of Middle East output at $25 per barrel, and Salameh (2004) predicted that prices are unlikely to rise above $50 per barrel; Fischer et al (2009) reveal continuous underestimations of demand in the EIA's (short term) forecasts.…”

Section: Introductionmentioning

confidence: 99%

Output adjusting cartels facing dynamic, convex demand under uncertainty: The case of OPEC

Wirl

2015

Economic Modelling

View full text Add to dashboard Cite

Understanding errors in EIA projections of energy demand

Cited by 26 publications

References 5 publications

Forecasting Residential Water Consumption in California: Rethinking Model Selection

Forecasting Residential Water Consumption in California: Rethinking Model Selection

Future U.S. Water Consumption: The Role of Energy Production¹

Output adjusting cartels facing dynamic, convex demand under uncertainty: The case of OPEC

Contact Info

Product

Resources

About

Understanding errors in EIA projections of energy demand

Cited by 26 publications

References 5 publications

Forecasting Residential Water Consumption in California: Rethinking Model Selection

Forecasting Residential Water Consumption in California: Rethinking Model Selection

Future U.S. Water Consumption: The Role of Energy Production1

Output adjusting cartels facing dynamic, convex demand under uncertainty: The case of OPEC

Contact Info

Product

Resources

About

Future U.S. Water Consumption: The Role of Energy Production¹