Loading Page..


Probability Elicitation, Verification, and Calibration

SDP Special Event Tutorial
Speakers: Shaun Comfort & Reidar Bratvold

Probability forecast verification and calibration are crucial in ensuring the accuracy and reliability of probabilistic predictions. Verification involves assessing the accuracy of the forecasted probabilities by comparing them to the actual outcomes, while calibration involves adjusting the forecasts to match the observed frequencies of events. Both processes are important for managing risk and uncertainty in a variety of applications, from weather forecasting, energy production, R&D investing, to financial modeling. Probabilistic methods are particularly useful for accounting for uncertainty, allowing decision-makers to make informed choices based on a more complete understanding of the underlying probabilities. Overall, probability forecast verification and calibration are essential tools for improving our ability to predict future events and outcomes and make better decisions based on probabilistic information.

In this SDP workshop, Professor Reidar Bratvold and Dr. Shaun Comfort will provide an overview of probability forecasting verification and calibration, along with worked examples from the energy industry and pharmaceuticals.
Zoom Webinar
Jun 14, 2023
08:00 AM - 10:00 AM
Hilda Cherekdjian
Respondent Email
List Price


Workshop Goals/Objectives

Provide users of probabilistic forecasts in Energy/Pharma/Technology an introductory background and set of techniques to:

  • Understand the purpose of Forecasting
  • Know the indications of a high quality probabilistic forecast
  • Understand and be able to apply Forecast Verification Techniques to simple probabilistic forecasts. 
  • Understand and be able to apply reference class re-calibration to common examples of poorly calibrated/post-hoc forecasts
  • Understand common pitfalls and best practices for producing calibrated forecasts



Reidar Bratvold | Professor of Petroleum Investment and Decision Analysis at the University of Stavanger.

At the Norwegian Institute of Technology, Reidar supervises graduate-level students researching decision analysis, project valuation, portfolio analysis, and behavioral challenges in decision-making.


Shaun Comfort | Neurologist and Principle Scientific Enablement Director at Roche-Genentech, Inc.

Shaun has 20+ years combined biopharma industry/regulatory experience.  His work focuses on combining machine-learning models and decision analysis techniques to evaluate Drug-Adverse Event causality and predict Project/Phase Probability of Success.



Who Should Attend? 

Practitioners applying quantitative Decision Analysis in Finance, R & D for Energy, Pharma, Medical Devices, and Technology.  If you just need a refresher on these topics or are new to probabilistic forecasting, you will find this SDP Special Tutorial useful!


   Email Print

Stay connected