Profitable systems vs curve fitting

Viewing 4 posts - 16 through 19 (of 19 total)

← 1 2

Author

Posts
05/30/2017 at 6:30 AM #36981 quote

Inertia
Participant

Master

Thank you Nicolas.

SharpeRatio please and Monte Carlo ?

Hard work Boss to maintain this forum active ! Congratulations !

Hope you had a relaxing sunny WE 😉

FYI: performances attached are not accurate (not enough granularity ;)) – Learnt from this forum 😉

MonteCarlo.png SharpeRatio.png

05/30/2017 at 8:12 AM #36994 quote

Nicolas
Keymaster

Master

I know Monte Carlo is already planned. Ok for Sharpe Ratio (already in my own wish list), any other things? I’d like to get the average MAE/MFE of all trades, the profit in points, the duration of winning trades and losing trades also (and yes thank you, I had a great sunny week-end 😉 )

05/30/2017 at 9:32 AM #37001 quote

Francesco78
Participant

Master

Hi Nicolas, we should have a way to assess the smoothness of the surface of the optimized variables.

Let’s say we have a system depending on x,y,z variables, we optimize it with respect to sharpe ratio and then we want to see how much the optimization results change if we change the variables x,y,z by 1%.

In other words we want an optimizer that gives information explicitly about the derivative of the optimized surface.

An other thing that will be very useful in my opinion is a tool that is able to draw the timeseries of the optimized variables, that is to say, suppose you have 3 year of history, then you optimize your system every 3 months in the past and so for every year you will have 4 data representing the optimized variable history every quarter.

It will greatly help in assessing the robustness of your system, together with the WF.

Many thanks

05/30/2017 at 9:50 AM #37005 quote

Nicolas
Keymaster

Master

Francesco78 wrote:

smoothness of the surface

You’ll get it with 3D charts, it is also planned to be featured in a future update.
Author

Posts