Dear all
I was wondering if any of you could help. In the field of data assimilation with numerical weather prediction we usually seek the mode of a posterior distribution conditional on a set of observations of the atmosphere which also has a distribution associated with it. Currently this is all assuming a Normal framework for all the distributions involved. i have done work to extend this to allow Lognormal variables and observations. the problem we have is this. In a recent proposal, which was rejected, we proposed to extend this work to allow for extreme event distributions, like hurricane intensity and extreme rainfall rates. One of the reviewers made the comment that we were concentrating to hard on the third and fourth moments rather than the 2nd.
So my question is this: How is the third moment related to the mode of the distributions? If you could refer us to some papers or a chapter in a book which may be available at CSU i would really appreciate it.
thanks for you time
steve fletcher
------------------------------------------------------------------------------------------------------------------------------------------
Dr Steven Fletcher
Research Scientist II
Cooperative Institute for Research in the Atmosphere
Colorado State University
1375 Campus Delivery
Fort Collins, CO 80523-1375
Ph: (001) (970) 491-8376
Fax: (001) (970) 491-8241
E-mail: [log in to unmask]
------------------------------------------------------------------------------------------------------------------------------------------
|