Max Henrion, PhD, CEO of Lumina Decision Systems, Los Gatos, California
Topic: Surprises and Black Swans in energy forecasts, past and future
Bio: Max Henrion is the Chief Executive Officer of Lumina Decision Systems, in Los Gatos, California. He has 30 years experience as a professor, practicing decision analyst, software designer, and entrepreneur, specializing in economic and environmental analysis of energy technologies. He is the originator of the Analytica software. He has been Professor at Carnegie Mellon, Vice President at Ask Jeeves (now Ask.com), and Consulting Professor at Stanford University. He is a Member of the Board of the Society for Decision Professionals. He is co-author of three books, including Uncertainty: A Guide to dealing with Uncertainty in Policy and Risk Analysis (Cambridge UP, 1990), and over 60 articles in decision and risk analysis, artificial intelligence, and energy and environmental policy. He has an MA in Physics from Cambridge University, M. Design from the Royal College of Art, London, and Ph.D. in decision analysis from Carnegie Mellon University. His work on “Rigs to Reefs: Decommissioning California’s offshore oil platforms” won the 2014 Decision Analysis Practice Award from the Society of Decision Professionals and INFORMS Decision Analysis Society.
Topic: Surprises and Black Swans in energy forecasts, past and future –
Retrospective review of the last 30 years of forecasts of energy quantities and prices in US DoE’s Annual Energy Outlook show that we are often surprised by actual values turning out quite different from our estimates. The same is true even of estimates of fundamental physical constants, as well as expert estimates of the future costs of nuclear, solar, and other low-carbon energy technologies. Actual values are too often many standard deviations from estimates. According to Nassim Taleb “Black Swan” events are unavoidable and inherently unpredictable. But might it be possible to reduce how often and how much we are surprised? I will present three different approaches:
- Apply error distributions from past AEO forecasts to obtain better calibrated estimates of their uncertainty.
- Explore the reasons for past surprises in energy forecasts to see if we can get earlier warnings of potential surprises.
- Conduct brainstorming with groups of experts to identify possible future surprise events that might affect the forecast quantities.
April 2015 Dinner Meeting Announcement
Presentation Material:
Presentation Slideshow – Surprises and Black Swans in Energy Forecasting, Past and Future