- This topic has 59 replies, 7 voices, and was last updated 6 years, 4 months ago by
.
-
Posts
-
Hey Shamasa,
I talked a bit more about that in the last course that I launched yesterday with the Top 10 USDCAD EAs. Really there is no precise answer to this question.
If you test the EAs too long on a Demo account, you will see the profits in the Demo account. Every strategy has its profitable periods and losing periods (you see that in the equity line of every strategy). So if you test it too long, you are in this profitable period for the strategy, and when you decide to move on live, you might hit that losing period.
Basically the more you test on the Demo, the closer you are getting to that losing period.
Petko, recently I use only the method you showed in the Top 10 AUDUSD course to avoid the Demo trading. I use the OOS.
After that, on the live account, I use a profit factor of 1.2 If the EA falls below that factor, I remove it from the live account.
It works great so far.
I am just watching the AUDUSD course, and I will implement the OOS. I see that many people wrote in the forum they are using it after the course and the update videos from Petko.
I use the OOS now all the time. I went a few times over this lecture and really liked the method. It saves time for Demo testing. Unfortunately I am not brave enough to skip Demo, so I test still there.
Hey guys,
Glad that the OOS helped. I use it more and more as well. I implemented it in some of the updates for the courses, so the new EAs that I launch every month should be better.
Desita, you really should not be brave in trading 🙂 You should be confident with the time.
Bob, glad to hear that you figured what rules to follow to work fine for you.
Hey, Rose! It is really. I use it more and more and I think I will include in more in my courses because the EAs are getting better.
I have missed that it is important to have acceptance criteria in the In Sample and the Out of Samle parts. Just now I realized that when the complete backtest is set, it is for all the period, but I saw now that is not…ahh..thanks Petko!
Hello Haliffa,
Actually the complete backtest is indeed for the whole backtested period. I will give you an example:
Let’s have a profit factor of 1.2 for the complete backtest. This does not mean that the profit factor in the In Sample or the Out of Sample will be above 1.2. Could be less, could be more. When you place for both and the whole period, then you will have your equity line smoother and at the same time, the Profit Factor will be above 1.2 for all.
Petko: I noticed in the list of indicators that the Donchian Channel was not part of my default indicators. I initialized the Donchian indicator and ran the USDCAD 15 minute reactor. The top three results used the Donchian Channel “bar opens below the lower band after closing below it”. The best result had a r squared of 97.52 and a profit factor of 1.4. The three best results just used different closing indicators. My questions are do you use Donchian Channels in your reactor runs? And if not what are your reasons? You have mentioned in your webinars that you do not pay much attention to which indicators are used in the list generated by the reactor. But when the top three all use the same indicator to enter a trade would this affect your decision?
Hello Donald,
Very good question here! I am still testing the results with this indicator. I noticed that it comes as a favorite in the Reactor results.
I always like to test a lot before including any EAs into the courses or my trading portfolio.
I would appreciate if you share some results with it.
Enjoy the rest of the Sunday!
This indicator works better on FSB pro with me because the position can open on the channel. In EA Studio opens the trades on bar open, which means it will be over/below the line. Which might be a lot further from the lines.. In FSB Pro it can open exactly on the lines.
Hello Bob,
Glad to hear you gave a good answer. What you say is correct but keep in mind that the bar opening is for every indicator with lines. Even for the MA Indicators. If there is a cross the position will open at the first bar opening. This way the cross is confirmed.
Hello,
I think that the bar opening is the only way to prove that there was a confirmed cross of any MA or another indicator.
Let’s say we have a Moving Average and the price is below, going up. In order to avoid the fake signals, we always need to see a new candle opening above the line/indicator. I am glad that EA Studio Experts work this way.
- You must be logged in to reply to this topic.