EVM Notes–February 12 , 2007

Attendees

• Bob Guth (Leader)
• Max Rudolph
• Tom Edwalds
• Cliff Angstman
• Bernie Rabinowitz
• Guests Walt Marsh
• Jeff Harper
• Ben Wolzenski
•  
• Secretary for call:
• Max Rudolph
•  
• Web liaison:
• Kevin Strobel

1. The minutes from the meeting of November 30, 2006 are posted on our group’s web site.

 

2. RMS update: ERMAP has funding for research projects and is actively seeking them out. The ERM Symposium will be held in late March, with slides and mp3 audio files of the sessions available at www.ermsymposium.org. ERMAP is also initiating an international affiliation led by Dave Ingram. Max also reported that Sara Sanford (executive director) and Michel Rochette (staff Fellow for risk management) are both no longer with the SOA.

 

3. Members of the Futurism Section presented information on how to utilize the Delphi method to develop model assumptions. The SOA published a research project in October 2005 on this topic, focusing in on Economic Variables. The project can be found at Delphi Study of Economic Variables Report.

    

   • The Delphi method does not come to a consensus with predictive powers, but attempts to approach a stable distribution of responses. In each round of the survey, respondents are asked to state their prediction along with their reasoning. Responses are shared with the participants anonymously to avoid having a few dominate the discussion through either loudness or credibility. Wikipedia has an article on Delphi at http://en.wikipedia.org/wiki/Delphi_method Cross-impact analysis is a method for revising estimated probabilities of future events in terms of estimated interactions among those events. We also discussed participation by experts, versus solid thinkers, and issues related to gathering information versus inference of results.
   • The Futurism Section is sponsoring a modified online predictive market, looking at mortality gains at issue as well as 15 and 30 years into the future. It will be published soon.
4. SOA influenza research project: The project will report the life piece before the spring meetings. Delphi studies were used to develop assumptions for insured mortality, where a consensus was reached, and asset variables where a consensus was not reached.

 

5. Copulas: Max suggested that Don Behan, recently retired from Georgia State, join the next call to discuss his recent research on practical use of copulas. He is open to this and will also be presenting this topic at the spring life meeting.

 

6. ERM Risk Poll: The group is interested in helping with the Risk Poll but not with the Risk Index project. ERMAP continues to discuss the project.