|
PureBytes Links
Trading Reference Links
|
Herman,
Thanks for posting your returns.
Seems to me that while the system reduced drawdown significantly, it
deteriorated system performance to a point of being significantly
worse than B&H.
It takes a lot of work to trade and I beleive the results should be
significatly better than B&H.
I have tried using stops based on ATR with positionsize fixed, and
even after optimization of stop levels, system performance suffers
considerably. In fact my results we better with a flat percentage
stop.
I am spending time in system development, but my current opinion is
that human judgement needs to be integrated with an entry signal. I
have not been able to program enough conditions to make mechanical
systems with high return, without severely limiting number of trades.
As negative as I may sound, I strongly beleive that posts like yours
benefit everyone who works at it.
While we spend a lot of time with system development and trying to
keep up with a barrage of AB developments by Tomasz, we must remember
that our focus is not to become programmers, but use our collective
skills to better our trading. Sometimes that objevtive seems to take
second place....
So I hope we see more posting on systems and trading in the future,
especially as AB enhancements will start slowing down and users will
have fewer questions as time goes on...
Ara
--- In amibroker@xxxx, "Herman van den Bergen" <psytek@xxxx> wrote:
> Hello,
>
> further to an earlier post by CS i added the van Tharp's
> ATR-based position sizing code to the StoRSI-BBP system we have
> been using on the list in several examples. I attach the backtest
> results for the N100 stocks. Typical results look like this:
>
> No Stops&PosSizing Tharp PositionSizing Ratio
> NoTharp/Tharp
> Range %Profit Max%DD %Profit Max%DD %Profit Max%DD
> QQQ 7/20/99-10/1/02 128.00% 34.00% 8.95% 2.00% 14.30 17.00
> N100 1250 bars 217.00% 85.00% 7.93% 6.04% 27.36 14.07
>
>
> If, like it has been mentioned on this list several times, the
> application of van Tharp's techniques are a matter of ruin or
> no-ruin then we should have at least one single person on this
> list of about 1000 who can to come forward with a practical
> example that works. My appreciation of this topic swings from
> great admiration one day to total dismay the next...
>
> Can somebody show where I went wrong and give a correct example?
>
> Best regards,
> Herman.
|