Vince,

 Thanks for responding to Christian. I had replied earlier and said that EVT 
is based on iid events, and is therefore not useful in weather, where there 
is so much correlation going on. Jitendra from AA raised the same issue 
yesterday.

Joe






Vince J Kaminski
08/22/2000 04:49 PM
To: Christian Werner/ENRON_DEVELOPMENT@ENRON_DEVELOPMENT
cc: Joseph Hrgovcic/HOU/ECT@ECT, Vince J Kaminski/HOU/ECT@ECT 
Subject: Re: extreme value theory applied to weathet  

Christian,

About two years ago I asked Joe Hrgovcic
to look tinto EVT applications. As far as I know
he is the only person in the group, in addition to me, looking at EVT.

Vince




Christian Werner@ENRON_DEVELOPMENT on 08/18/2000 07:00:42 PM
To: Vince J Kaminski@ECT
cc:  
Subject: extreme value theory applied to weathet


---------------------- Forwarded by Christian Werner/ENRON_DEVELOPMENT on 
19/08/2000 10:06 ---------------------------


Christian Werner on 19/08/2000 02:08:56
To: Vince.Kaminski@enron.com
cc:  

Subject: extreme value theory applied to weather


---------------------- Forwarded by Christian Werner/ENRON_DEVELOPMENT on 
19/08/2000 01:15 ---------------------------


Christian Werner on 19/08/2000 01:55:56
To: vkamins@enron.com
cc:  

Subject: extreme value theory applied to weather

Dear Vince,

Back in July, when you visited our Sydney office you mentioned extreme value 
theory. I am wondering whether research is looking into the application 
of extreme value theory to power and esp. to weather. In a recent news 
article it was highlighted that a trend in the industry towards t_max, t_min, 
etc. 

I am in particular referring to the news article below:

http://213.219.22.254/i-wex/resources/ef290600.htm

In the past we have observed a similar trend where customers are asking for 
t_max, t_min, or below or above precipitation structures. The choice in the
past has been the burn analysis on historical data. However, we are in 
particular interested in the extreme events, and the application of EVT could 
provide a meaningful tool for the analysis. 

Has the research group looked into the application of EVT to weather? EVT has 
a long history of application in hydrology (which would cover parts 
of the precipitation structures...). Also, research (esp. at ETH in Zuerich) 
is indicating the application of EVT to V@R....


Thank you! 


Regards,
Christian