Re: [amibroker] Artificial intelligence

 

Hi Howard,


Here's the general framework family:

Essentially, I have a classification of any security into three states, corresponding to attractors in a phase space. These correspond to two universal memory states (different shapes) and a random walk state. The two memory states take on specific shapes governed by certain parameters. 

I have an inference network which can assign a confidence level to the proposition that a given security is currently in a given attractor, for each attractor, or if it is in a phase transition. I have a price series for cash which varies with the specific distribution of confidence -- the lower the confidence in any attractors or the higher the confidence in a phase transition, the higher the value of cash relative to the security. I can get an aggregate price for cash for an entire portfolio by looking across all the element-specific price-of-cash estimates.

This is a model of the market as a dynamic system, with changes to memory being the unobservable variable which drives it. 

__._,_.___

Posted by: rosenberggregg@yahoo.com
Reply via web post Reply to sender Reply to group Start a New Topic Messages in this topic (24)
**** IMPORTANT PLEASE READ ****
This group is for the discussion between users only.
This is *NOT* technical support channel.

TO GET TECHNICAL SUPPORT send an e-mail directly to
SUPPORT {at} amibroker.com

TO SUBMIT SUGGESTIONS please use FEEDBACK CENTER at
http://www.amibroker.com/feedback/
(submissions sent via other channels won't be considered)

For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
http://www.amibroker.com/devlog/


.

__,_._,___


EmoticonEmoticon