Newsgroups: comp.ai.genetic
From: thasan@mayfair.demon.co.uk ("V.K. Thasan")
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!news.ultranet.com!news.sprintlink.net!pipex!peernews.demon.co.uk!mayfair.demon.co.uk!thasan
Subject: Help : GA Optimization
Organization: AIT
Reply-To: thasan@mayfair.demon.co.uk
X-Newsreader: Demon Internet Simple News v1.29
Lines: 21
X-Posting-Host: mayfair.demon.co.uk
Date: Mon, 20 Mar 1995 13:53:58 +0000
Message-ID: <795707638snz@mayfair.demon.co.uk>
Sender: usenet@demon.co.uk


I am currently working with a GA based analysis for financial markets.
I have the following problems:
        1. As we go along with each generations, the difference in fitness     
           decreases and then there is no improvement in fitness. Will this be
           rectified if I use some sort of fitness scaling? Any good scaling
           techniques!
        2. Are there any good estimates for p_crossover, p_mutate and optimum             values for max_num_gen  and num_populations.

Any help/suggestions are highly appreciated.

Vijay.   
-- 
**************************************************
* Vijay K. Thasan                                *
* Software Development                           *
* Arab Commerce Bank, London.                    *
* Tel: 0171 - 437 0483 (Off.)                    *
*      0171 - 373 5420 (Res.)                    *
*      0956 - 465 256  (Mobile)                  *
**************************************************
