|
PureBytes Links
Trading Reference Links
|
>The temptation to apply cutting-edge mathematical/engineering
>techniques to trading is hard to resist. For a good survey
>see: Deboeck: "Trading on the Edge." For another interesting
>viewpoint on the mental side of it, try: Kahneman et al:
>"Judgment under uncertainty: Heuristics and biases."
Interesting chapter in Market Wizards II about William Eckhardt. He was
working on his math doctoral dissertation just before jumping into becoming
a floor trader...still listed in the top 10 CTA's.
>The land baron down the road, who made his wealth in the markets
>once remarked, "engineers make the worst traders." I like to
>think of myself as a recovering engineer... :)
Funny, the "land baron down the road," from me is Ed Seykota: MIT
engineering grad and the "father" of computerized system trading
(Market Wizards I).
BW
>From: Jim Osborn <jimo@xxxxxxxxxx>
>To: omega-list@xxxxxxxxxx
>Subject: Re: Trading System Design
>Date: Thu, 10 May 2001 14:37:07 -0700 (PDT)
>
>Thomas Alexander <alexander_enterprises@xxxxxxxxx> writes:
> >What I would like to know is, what are GA and NN referring to?
>
>Ok I'll take this one... :)
>
>GA: Genetic Algorithm; an environment where the software can
>"evolve" its way toward a solution to some problem. Proponents
>have countered dismissive remarks such as "...but you wouldn't
>want to fly on a plane designed that way" with "I wouldn't
>want to fly on a plane that WASN'T designed that way."
>
>NN: Neural Network: simulation of synaptic connection architecture
>of the brain, that can be "trained" to recognize the solution to
>a problem.
>
> >And as far as I can tell, Easylanguage is a standard
> >linear programming language by definition, as compared to
> >object oriented or some of the various embedded AI language
> >variants. Did I miss something? Does Easylanguage have
> >"variants?" What's the "standard linear programming"
> >reference apply to?
>
>Linear Programming: a mathematical construct, often "linear algebra,"
>that allows many input variables to solve for some number of outputs.
>If you have fewer outputs than inputs, and the relational functions
>among them are all independant, and linear, you can find a unique
>solution.
>
>The temptation to apply cutting-edge mathematical/engineering
>techniques to trading is hard to resist. For a good survey
>see: Deboeck: "Trading on the Edge." For another interesting
>viewpoint on the mental side of it, try: Kahneman et al:
>"Judgment under uncertainty: Heuristics and biases."
>
>The land baron down the road, who made his wealth in the markets
>once remarked, "engineers make the worst traders." I like to
>think of myself as a recovering engineer... :)
>
>Have fun!
>
>Jim
>
|