Log in

Performance tests: What to expect

By JohnLast 2215 days ago Comments (2)
image
От 21 октомври 2011

Here I would like to add another study on the performance of Brain Trend tested with the expert. 

As this system us single entry and single exit, it is not the best monay management scheme, however it allows me to gauge the profitablility of the system and if the system matches well the market regime.

The tests were from the 05.10.11 until yesteday I think 17.11.11. 

The idea is that I am not interested in one setting alone. Really not at all, I am interested if the system has an overall profitability in all settings in this time frame. I am looking for a robust solution.

And the solution will be robust only if the system can solve the problems that are given my the market.

I mean I try to look at che charts with a different eye now, I am looking at the general patterns of volatility and fractal dimension. It is interesting that the market accts like changing its gears and it changes its behaviour. The key is to match this behaviour with your own tools whatever they can be.

The other important point  for Brain Trend is that it is either working or not working at all. and that will allow for a very precise risk management. Look at the drowdown this is what is expected from the system if the draw dows exceeds the average limits, we can make a more precise statistical computation of the average drawdown from all the settings is roughly 5 %, the maximum allowable drow down is 6.92 %.  

Comments

  • JohnLast 2048 days ago

    Here I will continue the tests with a little gap from the last tests:

    The expert is profitable except for the extreme lenght of the filter 110 and 120.

     

    image
    От 21 октомври 2011
  • JohnLast 2048 days ago

    It turnes out that we had a system that performed for a half of a year on almost all his possible settings.