PureBytes Links
Trading Reference Links
|
-----Message d'origine-----
De : Gary Fritz <fritz@xxxxxxxx>code-list@xxxxxxxxxxxxx
À : Code List <>; omega-list@xxxxxxxxxx <omega-list@xxxxxxxxxx>
Date : mercredi 4 août 1999 22:47
Objet : Pi trading system
A few days ago there was some discussion about the "Pi" trading
system from http://www.tradingsystems.freeserve.co.uk/. The general
consensus at that time was that, since it was based on P&F charts,
and TS doesn't handle backtests on P&F charts properly, it was almost
certainly bogus.
****
The consensus was two people among them, myself...
TS handles backtest on PF if you pay attention on how PFig charts are built.
TS displays HLC bars as they appear, and anyone should know that the close is known when the next bar is starting to build.
So anyone who writes a system on Pfig charts like this:
if something then buy on close;
...should not tell that Ts is guilty because you ask to it an answer from a point ( close) that is not known at this time ( that is absent from the horizontal PFig scale!).
At least they should have warned the user that do not understand technical analysis basis... ( also, do not forget to plug in you computer to make TS work as expected, and be careful when crossing the road).
The solution to build PFig systems that behave the same RT or not is to write:
If condition1 then buy at c + or - (boxzize*revsize) stop; ( +- according to the X O status of the latest PFig column}
=================================================================================
( With the exception of XO columns that occur during night gaps, not easy to address)
Yesterday I saw that the Pi vendor was offering a free trial copy of
his system, so I got a copy. I thought you all might be interested
in the results.
I applied it to a 0.25x3 P&F chart, ...
(snip)
... as you would have traded them, ***LOST*** $2920 for the
day.
snip encore
***
Was what I explained.
No need to buy a trial version for that.
I've gotta think the guy marketing this thing doesn't have a clue
that his system is bogus, or else he would never offer a free trial.
Anybody who runs the thing will immediately see it's worthless. You
have to wonder why **HE** never ran it. He's either too naive to
test his own systems in realtime, or he figures other people are too
stupid to do so.
***
Probably.
But he has sold one trial copy!
Sincerely,
-Pierre Orphelin
Représentant exclusif de Omega Research en France.
web: http://www.sirtrade.com
|