campbell
09-29-2008, 05:05 PM
If you get SOA emails, you've probably seen the announcement:
http://www.soa.org/research/risk-management/research-copula-phase.aspx
The Society of Actuaries' Risk Management Research Team is pleased to make available a research report on phase transitions and copulas. This report, authored by Michael A. Ekhaus of Gibraltar Analytical, LLC, has two primary objectives. The first is to incorporate phase transitions into a copula framework, and the second is to apply this to calculate the probability of defaulting in the context of a portfolio of correlated mortgages when a contagion effect exists.
Paper here:
http://www.soa.org/files/pdf/research-2008-copula-phase.pdf
I think this "contagion phase transition" is a very timely idea, and can be helpful beyond mortgage modeling, of course. I've just started reading the paper, so I don't know how practical it is.
http://www.soa.org/research/risk-management/research-copula-phase.aspx
The Society of Actuaries' Risk Management Research Team is pleased to make available a research report on phase transitions and copulas. This report, authored by Michael A. Ekhaus of Gibraltar Analytical, LLC, has two primary objectives. The first is to incorporate phase transitions into a copula framework, and the second is to apply this to calculate the probability of defaulting in the context of a portfolio of correlated mortgages when a contagion effect exists.
Paper here:
http://www.soa.org/files/pdf/research-2008-copula-phase.pdf
I think this "contagion phase transition" is a very timely idea, and can be helpful beyond mortgage modeling, of course. I've just started reading the paper, so I don't know how practical it is.