ARA Boosts Forecasting Methods to Improve Predictions, Intelligence GatheringNovember 2011
Forecasting ACE seeks a broad range of opinions from volunteers with diverse educational, professional and vocational backgrounds.
“Participants can input forecasts daily, weekly or monthly. It is up to them,” said Dr. Ed Merkle, a research partner at the University of Missouri. “They will be asked to answer questions on topics that interest them and then will receive updates on what others are predicting. They will also receive advice on how to make better forecasts.”
Get more information about the project, test your forecasting skills, and learn how to participate at www.forecastingace.com.
Every day intelligence agencies rely on informed and experienced experts to provide forecasts on world events – which in turn shape governments’ actions on the world stage. But whose predictions should rise above the crowd? That’s the question that ARA is tackling in its new ACES project.
ARA is leading a multidisciplinary research team, sponsored by the Intelligence Advanced Research Projects Activity (IARPA), to develop superior forecasting methods that aggregate experts’ opinions on world events. ARA has teamed with top experts in judgment and decision-making from prestigious universities across the U.S. to identify the traits that make people particularly adept at producing accurate forecasts.
U.S. intelligence agencies are seeking a more powerful prediction capability that combines the opinions of many experts to make superior forecasts on world events. ACES provides that prediction capability using new crowdsourcing methods that take advantage of advances in judgment and decision-making, mathematical processes for combining opinions and communicating forecasts to decision-makers.
ACES moves beyond the phenomenon known as the wisdom of crowds and explores new methods for combining the opinions of many widely dispersed experts into predictions that are far superior to any individual’s forecast. ACES provides a virtual platform to engage participants, allowing individuals to identify topics of interest, engage with other experts and compete to make the best predictions.
“It has long been known that simply averaging a large set of independent judgments creates an estimate that is usually more accurate than every individual judgment in the group,” IARPA ACE Program Manager Jason Matheny said.
The aggregation techniques that ARA’s research team is exploring will identify experts in a crowd of forecasters and weight their judgments as more predictive of the true outcome of a forecasting problem. Using the ACES platform, experts will offer individual, independent and anonymous opinions and judgments, and the team will use our innovative aggregation methods to produce collaborative forecasts – ultimately leading to higher fidelity predictions.
“Some people know important details that can make the future predictable in many cases. They often do not share their insight with others. Our project is designed to find out what people know, have them share this knowledge with others, and ask them to make a prediction based on what they and others know,” ACES Principal Investigator Dr. Dirk Warnaar of ARA said. “In essence, everyone is an expert. You may be very good at digging up, recalling or combining relevant pieces of information that all contribute to making an informed prediction. We do not yet know which qualities make you a good forecaster, but we hope you’ll help us find out.”
Please visit www.forecastingace.com to participate.