|
PureBytes Links
Trading Reference Links
|
Fred,
We have bantered about what constitutes a 'drawdown'.
But my main question was...
Is your Equity Curve calculation for portfolio trades based upon
chronologically ordered trades that we would experience in real life,
or is the calculation based upon the artifical sequence of trade lists
that our backtest execute against?
Phsst
--- In amibroker@xxxxxxxxxxxxxxx, "Fred" <fctonetti@xxxx> wrote:
> I understand your need to do the evaluation and as I've stated
> unfortunately there is insufficient information even from the AB
> trade list to do this. If you hold a trade for a month and for that
> trade you bought at 100 and sold at 90 what you see is a 10% loss but
> if in between the entry and exit date the stock in question reached
> 120 before going to 90 you have a 25% drawdown. This information is
> NOT available from the trade list.
>
> --- In amibroker@xxxxxxxxxxxxxxx, "phsst" <phsst@xxxx> wrote:
> > From Fred:
> >
> > >Keep in mind when exporting available numbers even for each trade
> from AB
> > >that this does NOT allow for showing true drawdowns.
> >
> > and
> >
> > >In short, the point I'm trying to make here is that when
> developing,
> > >testing and optimizing systems IMHO the way to do so is using full
> > >compounding as to do otherwise shows erroneous results that can
> > >easilly lead one to think a system is better then it really is in
> > >terms of the negatives and worse then it really is in terms of the
> > >positives.
> >
> > Question for Fred and for TJ:
> >
> > In my post "RWT - Positionsize" I explained the need to analyze
> > backtest generated trade data in chronological order, matching
> actual
> > trading environments.
> >
> > Regarding how the Equity Indicators are calculated, is Equity
> > calculated in:
> >
> > 1) Alphabetical or Watch List Sequence
> >
> > or
> >
> > 2) Chronologically for all trades
> >
> > And regarding the compounding issue... does compounding occur on the
> > entire portfolio chronologically (the sequence in which you would
> have
> > actually made the trades) or is compounded in an issue after issue
> > after issue fashion. If it is not chronological trades using the
> > entire portfolio, then it is useless and misleading information.
> >
> > If I am right, then fixed positionsize is the only *potential* way
> to
> > analyze backtest output further. (And if I'm wrong then PLEASE let
> > someone explain why chronological analysis isn't necessary).
> >
> > Phsst
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Make Money Online Auctions! Make $500.00 or We Will Give You Thirty Dollars for Trying!
http://us.click.yahoo.com/yMx78A/fNtFAA/i5gGAA/GHeqlB/TM
---------------------------------------------------------------------~->
Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|