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 -~----------~----~----~----~------~----~------~--~---
