Thanks Eugene. That helps.  I also have been looking at percent
winning trades
vs percent loosing trades because its straightforward.  Do you think
that overlaps with
Profit Factor, PI, + Kelly?

Based on Profit Factor , PI , Kelly and %winner %looser. I haven't
been able to
beat the best JBT sample strategies for the dataset I have.  (saving
my lunch money
to buy more data) .  Does Equalizer3  perform really well for the full
year dataset?
If so why not trade it?



On Jan 7, 10:30 pm, nonlinear5 <[email protected]> wrote:
> I'll make a few points:
>
> 1. The "net profit" is probably the worst measure of performance,
> because it does not account for risk whatsoever. Max DD is also
> misleading, because it relies on the certain sequence of trades which
> is really irrelevant. In a Monte Carlo simulation, the trades are
> shuffled. On this shuffled sequence of trades, Max DD will be
> different every time, yet the trading strategy is the same.
>
> 2. I use Profit Factor and PI almost exclusively, and Kelly once in a
> while. All three (Profit Factor, PI, and Kelly) normally go together.
> That is, the strategy with a high PI tends to have a high profit
> factor and high Kelly.
>
> 3. The larger the number of trades and the smaller the number of
> strategy parameters, the more statistically significant are the
> results. Also, PI by its very definition acts as a statistical
> significance test.
>
> 4. To evaluate the robustness of the strategy, look at its performance
> when the best strategy parameters are changed by 15% or so. Is the
> strategy still profitable, or has it lost all of its merits?
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"JBookTrader" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/jbooktrader?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to