Forecasting Methodologies and their Implications
A common fallacy afflicting even sophisticated analysts is that they seek immutable laws of human behavior, much like the physicist discovers physical laws through experiment. Such generalizations about human and economic systems often fail because these systems are adaptable in ways that physical systems are not. Policy choices affect how the future unfolds, and parameters that embody historical behavior are bound to lead us astray whenever a forecast relies on those parameters to forecast far into the future.
Heavy reliance on statistically-derived historical parameters is what I call "The Big Mistake", and it is one that afflicts both "top-down" macro modelers and "bottom-up" engineering modelers. Assuming that human behavior is immutable will inevitably lead to errors in forecasting the future, no matter which kind of modeling you do.
Alan H. Sanstad
The analysis of tax credits for efficient equipment that we conducted for the Clinton Administration in Autumn of 1997 explicitly avoided The Big Mistake by using empirical data on rebate response to characterize three effects of the tax credits: a direct price effect, an announcement effect (the existence of a rebate has an effect independent of the size of the rebate), and a learning effect (based on cumulative production experience).
The Clean Energy Futures study avoided The Big Mistake by explicitly changing discount rates, technology costs, and other behavioral parameters in the NEMS model to reflect a scenario in which aggressive programs and policies reduce carbon emissions in the U.S.
Koomey, Jonathan G. 2000. "Avoiding 'The Big Mistake' in Forecasting Technology Adoption." Presented at the Energex Conference, Las Vegas, NV, 23-28 July 2000. LBNL-45383. (A modified version of this paper has been submitted to Technological Forecasting and Social Change). Abstract | 32K PDF
John A. "Skip" Laitner. 2000. "The Information and Communication Technology Revolution: Can It Be Good for Both the Economy and the Climate?" Discussion Draft, last modified 31 January 2000. 33K PDF
International Project for Sustainable Energy Paths: This non-profit organization analyzes the costs of reducing carbon emmissions and addresses the policy issues surrounding the design of sustainable energy strategies. The analyses the group produces are among the few that systematically avoid The Big Mistake.
Forecasting Principles: This site, created by Professor J. Scott Armstrong at the Wharton School of Business at the University of Pennsylvania, contains data, software reviews, software programs, forecasting literature, and other forecasting-related information for researchers, practitioners, and educators.