Optimise an EA and get the best of the optimised ones. - page 37
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Launched
But, as I don't have this pair, it's a long and painful story to download and distribute to agents...
I read here an interesting indicator for evaluating the TS - the time it takes for the TS to come out of a drawdown.
I use this indicator as a reason for withdrawal from bidding. (Long Max Wait).
When optimizing - data for its definition is laid down in the code. Each TS knows how many deals were on the test period, and knows how much maximal amount of deals was needed to drawdown. Accordingly, your figure can also be calculated.
39 rgcodes.
Current TCs for recycling:
I place AUDJPY EMATrendSAR
Optimization period 28.04.17 - 28.04.18, forward from 28.09.17
I use this indicator as a reason for withdrawal from bidding. (Long Max Wait).
When optimizing - data for its definition is laid down in the code. Each TS knows how many deals were on the test period, and knows how much maximal amount of deals was needed to drawdown. Accordingly, your index can also be calculated.
So, perhaps it is not deals that should be considered, but the time. I think we mean that the strategy stopped working due to the change of the market phase and we should understand how long we will have to wait for the market phase to return. Market phase - let it be trivial trend/flat. Time is important, because in theory it should be compared with other instruments - deposits, bonds, real estate for rent and so on.
So perhaps it's not the deals that should be counted, but the timing.
I have had the opportunity to do so, but it seemed to me that trades are a more accurate representation of the essence. In principle, the number of trades per time unit in each TS is quite stable. So there is not much of a difference.
And about "waiting for the phase to return" - this is, in my opinion, the wrong direction. Why should we wait for the phase to return to this TC, if a lot of TC - is now in the phase of earnings? Let's switch to them and don't bother.
Putting it to recycling
I have had the opportunity, but it seemed to me that the trades were a more accurate representation of the essence. In principle, the number of trades per time unit in each TS is fairly stable. So there's not that much of a difference.
Well, in the case of the available TS, maybe, but as a general indicator it might be interesting.
And about "waiting for the phase to return" - that seems to me to be the wrong direction. Why would we wait for the phase to return in this TC, if a bunch of TC - now in the phase of earnings ? Switch to them and don't bother.
Who cares, anyway, part of the time we spend to recognize the change of this phase, and it may happen that after recognition and termination of the phase will return, so this indicator can also tell us how much time is needed to recognize and change the behavior of the EA (replacement with another and so on). You can take not the highs, and significant drawdowns and derive the average, which may also be useful for analyzing the TS.
Current TCs for recycling:
I place a CADJPY EMATrendSAR
Period 29.04.17 - 29.04.18, forward from 29.09.17
recently there was a script about a long screenshot, it is not bad to just euro, euro pound and pound yen, 3-week chart on h1 with these indicators to lay out, there will immediately find what works in segments
it is important to see or modify in one of these indicators a real reasonable movement, and use entries on the basis of the nearest volatility
why do many people say that you need to periodically optimise the EA and that it works on a certain interval? because the system works on unsubstantiated data, for example
___
There is an indicator, the opening price is higher than it - buy, if it is lower - sell (it doesn't matter which indicator, any), based on history, such systems will show different temporal growth and loss, and spreads and sliding will further reduce the chances,
what it follows, grid system should be very tricky, and a lightly optimized system will be averaged and lose at least because of the spread and slippage
___
the point is that you need 1-2 clearly justified indicators on the chart, not an optimised bunch (c) Fast235)