forex software

Create and Test Forex Strategies

forex software

Skip to forum content

Forex Forum

Testing and Trading Forex Strategies

You are not logged in. Please login or register.

EA Studio offline for maintenance on July 24, 2019, at 00:00 UTC

A two-hour window has been allocated for the maintenance, however, the actual downtime should be much less.
This is mandatory security maintenance from the hosting company we use.

Forex Forum → Forex Strategy Builder Professional → Monte Carlo - too good results

Pages 1

You must login or register to post a reply

RSS topic feed

Posts: 14

Topic: Monte Carlo - too good results

Hello Mr. Popov,

when I analyzed one of my DAX M15 strategies, I noticed that Monte Carlo calculated much better results than the original strategy. It behaves similarly also in other timeframes and even with some other underlyings. What do you need to send from me to check this behaviour?

Post's attachments

basic strategy.gif
basic strategy.gif 14.7 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

Re: Monte Carlo - too good results

Here the Monte Carlo

Post's attachments

Monte Carlo.gif 133.43 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

Re: Monte Carlo - too good results

Monte Carlo calculated much better results than the original strategy

This is not a reason for an issue. Monte Carlo changes the market or trading behavior and the simulated conditions may be more suitable for your strategy.

You can check the simulations one by one and to determine which one of them produces better results.

Check also whether all the results are better or only some of them.

Check the spread and slippage values. The settings in the Monte Carlo options may be lower than what you have in the backtester.

Re: Monte Carlo - too good results

What is the performance before running the Monte Carlo?

Re: Monte Carlo - too good results

Set Indicator parameters variations to minimum and see the performance.

http://s28.postimg.org/s4qdk26jd/screenshot_828.jpg

Re: Monte Carlo - too good results

The performance before running Monte Carlo is in my first post.

Attached is the second Monte Carlo run (again 20 calculations performed) with indicator parameters set to minimum.

Post's attachments

Monte Carlo 2.gif 26.41 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

Re: Monte Carlo - too good results

What is the performance  shown on the Monte Carlo chart before running the simulations?

Re: Monte Carlo - too good results

I think I succeed in reproducing the issue. According to my test it appears only when "Randomize indicator parameters" option is used. Are the tests normal using other simulations?

Re: Monte Carlo - too good results

Here it is.

Post's attachments

Monte Carlo 2 - before running simulation.gif 46.57 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

Re: Monte Carlo - too good results

Yes, other tests seem to behave normally.

Re: Monte Carlo - too good results

Ok. Thank you for the report.


http://s28.postimg.org/gtnpvozo9/screenshot_829.jpg


I'll find and #@@#!! it.

Re: Monte Carlo - too good results

Thanks also to you for the prompt support and for the great platform you have created. smile

I have just one additional wish concerning Monte Carlo: Very frequently, after I run the simulation with "Randomize indicator parameters", I have then big problem to find out what particular parameter is a bottleneck. Would it be possible to incorporate a "post-Monte Carlo" analysis tool that would help to find quickly the critical parameter - in the same way as the Strategy Analyzer tool in the original FSB?

Re: Monte Carlo - too good results

Yes, certainly.

I have it in my To Do list for the next major release

Re: Monte Carlo - too good results

The bug is fixed.

Thank you for the report. However the backtest of your Monte Carlo simulation is correct. The tester simple set the default list params. smile

Posts: 14

Pages 1

You must login or register to post a reply

Forex Forum → Forex Strategy Builder Professional → Monte Carlo - too good results

Similar topics in this forum