@ jusmih1,
I just finished optimizing the V1.2 on 200,000 units.
this version will be V1.2.3
Whatever the version we see a difficult period at the end of 2017, this period is shorter on V.2.3
Good luck !
@ jusmih1,
Last test: successful OOS validation with Walk Forward module.
V1.2.3 is robust.
Hi Gertrade,
thanks for taking the time to look at and test my version of the strategy. Its much appreciated, i guess its down to how all the versions perform this week, fingers crossed!
Thanks
justin
@GraHal thanks for taking the time to reply to my posts earlier, #62245.
I am still far from being a coder (haha) but certainly have a better understaning of how its all made up.
Hope you all have a good week trading!
Thank you all.
cheers
Hi every body,
I give you the first results of the day: on all the versions that I published, only 2 versions are profitable tonight.
This is the V2 and the V2.2. Only the V2 has been profitable today. The V2 is better than the V2.2 daily since 05/02.
I have no EURUSD Mini results, I do not have it in the library on my demo account.
These days, we have seen that it will divide my results by 20 to make the comparison with the EURUSD Mini.
I am waiting for your comments and results to compare.
Hi all,
I have the following proof of differences between Real Time Backtest and ProOrder Real Time.
So do not blindly trust the backtests that are failing.
I had doubt for several days, my backtests were pretty good and the results ProOrder mediocre.
I reversed the orders in my strategy and there I became profitable on ProOrder.
Who can help me ? I feel helpless in the face of this problem.
Be careful, do not reverse your positions on your strategies, it is not necessarily the solution, I became profitable for a short period.
On the other hand, it is necessary that this problem is regulated at the level of Backtests.
The use of TF below the minute should accentuate this phenomenon.
@Gertrade … Re EURUSD and short Timeframes … isn’t it likely that the equity curve is curve fitted on backtest at a fixed spread (input by user) and on Live trading EURUSD has a variable spread?
Also what do you mean when you say … Real Time Backtest? How can you have … Real Time Backtest?
Only way data is real time is actual trades Real Time on Live Platform?
Next nearest to real time is actual trades Real Time on Demo Platform?
Cheers
GraHal
Take a look at your orders list and their executions times in your demo account. Entries should differ a bit and may explain these differences you spotted.
@ Nicolas,
Thank you for devoting a little of your time.
I do not find in the list of orders their execution time.
Do you have a method to do this?
@ Grahal,
I think between 23h and 01h the spread varies, after 01h am we are closer to 0.6. This necessarily has an influence on the equity curve.
I used the term real-time Backtest to define that once the Backtest started, I would look at its evolution in time.
The real real time belongs to the platform live with the market but the demo platform must approach and be closer.
I hope that my answer will suit you. As a Frenchman, I make a lot of effort to understand your English.
Just hit CTRL+O, you’ll get the entire orders list.
@Gertrade is it any easier understanding my English if you right click and select translate to French?
And maybe it might be easier for you if you type in French and then right click, select English and then copy and paste the english onto the page??
Let me know if my idea is any good??
Cheers
GraHal
PS Backtest is the process of running the Optimiser to get best results etc. Then when it is finished running you get to see an equity curve displayed.
Please correct me somebody if I am using incorrect terminology??
@ Grahal,
Thank you. It is a good method of translation.
Hi guys,
Maybe I’m doing something wrong…I have imported all the stategies into my prt to check with probacktest wichone could be the better one and if I can…try to improve it.
Probacktest with 100000 units and 5 min timeframe on eurusd or eurusd mini….noone leaves for more than 2 mounth.
I’ve tried also with 28 sec timeframe or 18, but…same result…all of them dies after few times…
Am I doing something wrong?
