|
PureBytes Links
Trading Reference Links
|
The game teaches that even with a positive expectancy, human nature
will cause you to lose if you don't know how to position size. As
was demonstrated by Ralph Vince's study of 40 Ph.Ds where only 2 came
out ahead in the end. To find the optimal amount to risk on each
"trade" for say, the % risk model, just set up a simulation of 1000
trials of X trades for each % step and compare the cumulative
probability curves.
Using a bag of marbles to sample from with replacement as a
representative sample of the imaginary "larger population of marble
trades," is a good preparatory drill and *almost* analogous to using
historical test results as a representative sample of future trades.
The difference being, of course, that it is extremely difficult to
determine which historical test results are most representative of
the future. But if you have a set that you believe are
representative, you can do the same position sizing drill that you
did with the marbles.
Mark
--- In amibroker@xxxx, "Richard Alford" <richard.alford@xxxx> wrote:
> If the process is random (as the game likely is) there is no such
thing as a "streak", specifically it is a Markov process with no
history feedback. It would be interesting to understand just what
Tharp's game teaches.
>
> Some think/hope/assume/pray/etc, btw, that there is information in
the time history of an security. A "system" that exploits this
history and get to conditions which are favorable may indeed have a
"streak".
>
> www.bjmath.com has some interesting articles. The BJ is BlackJack,
which has well know properties that can tilt the odds towards a
player
at times.
>
> Cheers
>
> Richard
>
> ----- Original Message -----
> From: E Winters
> To: amibroker@xxxx
> Sent: Friday, June 28, 2002 6:02 PM
> Subject: Re: [amibroker] Trading system development and
simulations
>
>
> David,
> Van Tharp has a demo of his trading program on his web site. If
you agree to get a free copy of his newsletter and register (he does
send announcements about his courses to registered users) you can
download it and try it out. Look for the free trading game box on
the
main page at www.iitm.com
>
> One theory is based on the theory of runs. When you are on a
winning streak you should risk more per trade until you have a loser.
Try adding to winning streaks and see how it improves trading.
> Regards,
> Ed
> ----- Original Message -----
> From: David Holzgrefe
> To: amibroker@xxxx
> Sent: Friday, June 28, 2002 5:41 PM
> Subject: Re: [amibroker] Trading system development and
simulations
>
>
> AL an interesting article , Has anyone tried his trading game
??
what did they learn from it
>
> David
> ----- Original Message -----
> From: Avcinci
> To: amibroker@xxxx
> Sent: Saturday, June 29, 2002 6:32 AM
> Subject: [amibroker] Trading system development and
simulations
>
>
> Here is a very interesting article I got from Tharp's forum
by
Frank Gallucci. It is about trading system development and
simulating.
It's a good read: http://www.iitm.com/Weekly_update/simulations1.htm
>
>
> Al Venosa
>
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of
Service.
>
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.372 / Virus Database: 207 - Release Date:
6/20/2002
>
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of
Service.
>
>
> Yahoo! Groups Sponsor
> ADVERTISEMENT
>
>
>
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of
Service.
|