wilmott.com

Hi,does any body has proficiency in weather derivatives ?Can you suggest a reasonable distribution for rainfalls ?

Often weather derivatives are locked to some weather trait measured at a specific place or weather station. Maybe you can find historical data for a place or weather station relevant to your case and draw some conclusions about the distribution from that data./TGN

Why just a distribution? Write some weather forecasting algorithms and trade based on that, there is lots of literature, especially with regard to rainfall.

Here is an aticle for forecasting temp for weather derivatives:http://www.econ.upenn.edu/~fdiebold/pap ... rint.pdfHe talks about density forecasting as well.

There's a book by Jewson and Brix called Weather Derivative Valuation that might be helpful. Unless you're talking about long-dated derivatives, you'll need to model a lot more than just historical distributions; you'll have to get into weather forecasting too. Good luck.

Weather Forecasting for Weather Derivatives, by Sean D. Campbell and Francis X. Diebold is a another good paper. Link:http://opim-sun.wharton.upenn.edu/risk/ ... -31-SC.pdf

QuoteOriginally posted by: mshiversWeather Forecasting for Weather Derivatives, by Sean D. Campbell and Francis X. Diebold is a another good paper. Link:http://opim-sun.wharton.upenn.edu/risk/ ... 31-SC.pdfI think you missed my post

i remember hearing that the tweedie distribution is a common distribution for rainfall. its a compound sum probability model given byS=X1+...XNwhere Xi is gamma-distributed and N is poisson distributed. It's probability distribution is easily recovered through the FFT. I guess the interpretation here is N is the random number of rainy days over some time horizon and X is the random amount of rainfall on such a day.

Yeah, it's easy. The distribution is everywhere except my farm!