In search of the sacred 'grail'... - page 3
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
Here another thing is that by using several indicators we have too much information. For example, let's assume that the Force Index indicator gives a trade shift of five percent (probability of positive outcome is 55%), as well as assume that the AO indicator gives a trade shift of 7%. It would seem that using these two indicators, we should gain a trading advantage of 12% or at least close to this value, but it can't be. Why? The Force Index considers AO (!), because both indicators are based on the mean values of prices for a certain period, i.e. in this case they both contain the MA indicator! Even if we use two indicators that are not based on MA (which is rather difficult to do, because they are almost all based on averages), we still get redundancy. The reason? The reason is that all technical indicators already take the market price into account. I think it is senseless to use more than three or four indicators. The first two of them will take almost everything into account. This paradox from the field of cybernetics I cannot prove yet, but practice speaks in favor of this assumption.
Every additional indicator, and indeed any additional parameter, acts as an additional filter and
naturally, reduces the number of deals.
And since the market is poorly predictable, reducing the number of trades leads to over-optimisation and adjustment to the story.
Therefore, it's better not to supplement the basic strategy with additional indicators, but allocate additional indicators
into a separate strategy, optimize them separately, and use them together with the basic strategy.
This will create a hedging effect, i.e. reduced drawdown with the same or slightly higher profit.
Managing the size of TP and SL, setting the maximum soft Money Management in combination with limiting the number of trades
it is possible to make almost any unsuccessful strategy profitable over a long time frame, e.g. 10 years, in a day.
The most unsuccessful strategies give profit equal to bank deposit interest.
Unsuccessful of such strategies, usually becomes quite obvious by visual analysis.
The most successful - up to 100-150% per annum (on average over the entire testing period).
This 1-2 million dollars over 10 years, with start-up capital of the order of 2000-3000 $
Besides, this approach allows to play during the day, which for me, for example, is important :-).
I can't figure out why we have a feminine form of GRAAL )
maybe they meant to write In Search of the Sacred "GRAIL"...
i think it makes sense to herd different turkeys in the same flock of only their own production
А так как рынок слабо предсказуем, то снижение количества сделок ведет к переоптимизации и подгонке под историю.
Why is that? What is the correlation between the number of trades and the history fit?
It will create a hedging effect, i.e. reduced drawdown with the same or slightly higher profit.
And if both strategies will start losing money? Then instead of hedging effect it will have bankruptcy effect. I think there should be a hedging strategy, but it should be based on completely different market properties. For example, if the main strategy trades in a trend, the better is the result, the stronger is the trend, the hedging strategy can be based on flat trading. That is, it is not the indicators used that are important, but the market model.
Why is that? What is the correlation between the number of trades and history matching?
If you use reinvestment using the simplest formula, for example,
And then include the parameter Maximum Risk per trade MaximumRisk,
you will immediately see trades start to accumulate in the most profitable places, e.g,
SELL at the very top, BUY at the very bottom.
Between the very top and the very bottom, the deals will practically disappear and the MaximumRisk will be
tends to 1.
But here is a problem. These extrema occur once in 3-5 years and all the time they are in different places.
So what do we do in between - wait for the extremum?
Of course playing only on the extremums does not guarantee maximum profits.
And what if both strategies start to fail?
This is why we should choose strategies thoughtfully and not bet similar ones in a hedge, for example,
do not put three strategies based on intersection of iMAs of different periods.
You should e.g. have one based on iMA crossing, one on channel breakout and one on statistical prediction.
Also, as I said before, set extremely soft Money Management and adjust
TP and SL, so that TP is much bigger than SL, for example TP=1000 and SL=100,
Set in the optimizer a limit on the number of continuous losing trades of 10.
I think there should be a backup strategy, but it should be based on completely different market properties. For example, if the main strategy is trading in a trend, the better is the result, the stronger is the trend, the hedging strategy can be based on flat trading. That is, it is not the indicators used, but the market model that matters.
It is very difficult to do because it is difficult to make an indicator that predicts the future market behavior.
And the concept of trend and flat is very relative and depends on the chart period.
On H1 it may be flat, while on D1 it may be in a strong trend.
But we can control the type of the strategy, and we should start from there.
You just shouldn't be greedy and try to earn $1 million a year with $100.
You have to work out a long-term strategy and come back to the short-term one later,
in the Maldives.
If you're 80 years old, you might not even think about a long term strategy.
If we have time to drain the rest :-).
"omitting" indicators, Figar0 offers you, Hoper23, to work directly with bars, i.e. directly with price information, without intermediaries
I am still far away from complete refusal of indicators) Not exactly what I wanted to say, although I myself use indicators very rarely and always very cautiously, only after a deep understanding of their essence ...
But it's up to everyone, you can use indicators, if it's easier for someone, but their use should be sensible, i.e. you have to understand what kind of indicator is used in the system and what it shows. And what does the first post say? "Give me some indicators, let's try them, the grail is coming. The Expert Advisor has been showing profit for half a day"? Where is the sense here? The chance to get something working is equal to chance.
Like I said, this way some people have to go by themselves, precisely because half a day of profit will not give any arguments)
и подберите TP и SL, так чтобы TP было значительно больше SL, например, TP=1000, а SL=100
This is definitely not working. All other things being equal, the high level of profit is compensated by the low probability of execution. That is, if at SL=TP, the probability of execution of both is 50/50, then at setting TP double the SL, the probability of TP execution will also double and reach 25%, respectively, the probability of SL execution will increase to 75%. These are the fundamental rules of game theory, clearly proven by testing of random EAs.
Not exactly what I wanted to say, although I myself use indicators very rarely and always very cautiously, only after a thorough understanding of their essence...
But it's up to everyone, you can use indicators, if it's easier for someone, but their use should be sensible, i.e. you have to understand what kind of indicator is used in the system and what it shows. And what does the first post say? "Give me some indicators, let's try them, the grail is coming. The Expert Advisor has been showing profit for half a day"? Where is the sense here? The chance to get something working is equal to chance.
As I said, some people must go this way on their own, precisely because half a day of profit will not give any arguments)
My experience is that it is best to work with both indicators and bars,
for example Close[n]-Indicator(n).
And to analyze not the absolute value, but the range of values.
A fan in this interpretation is a gap or several gaps.
I.e. analyse Min_1<Close[n]-Indicator(n)<Max_1,,, Min_n<Close[n]-Indicator(n)<Max_n
Analyzing this information we eliminate lags of indicators and get a more adequate picture of
picture of market behavior.
However in this formula Close[n] can be replaced by iMA with a small period or iMA on a smaller timeframe.
"Gratuitous" does not smell here, it smells of ridiculousness and tinkering to "work" on history. The approach of bumping indicators like this will achieve nothing. The profit of the system should be in the ideas, it should be reasonable at the design stage, and at the stage of construction - adjustment and tuning. It is possible to express everything or almost everything with a multi-period set of indicators, it's just a question of how to interpret them. In general, I think that 90% of indicators, and certainly almost all of the classical indicators, are a relic of manual trading, for automatic trading we need other tools. Well, this is your business, I will not interfere with my own ideas.) If you do not understand what is going on in this situation then you may fall down on your own tramp, and it will lead you to enlightenment.)
I absolutely agree, the main thing is the idea!!! And then how to implement it is a matter of technique, and then the standard indicators are unlikely to help, or help ... But only in a secondary quality. I've done away with indicators for a long time now, because I realized they were useless. Or rather, in my TS I use only one indicator, just for convenience .... not to calculate the average manually :)