I will write an advisor free of charge
Here's an idea. I've already put it up for discussion - no one has responded - it's summer - everyone is on vacation or in jail.
Here's an idea: introduce price movement in decibels and set zero or zero line on small noises or flat and at a sharp move, for instance by 10% to set an order, then depending on the strength of the average tick calculated from the history, add +51%, set 0 again, and it's like that again in steps. The problem is to catch a candle or daily trend with at least 30-50%, it's like that. Well, a reversal: depending on the sine of the angle, it has gone higher than one by 10-20%, which means a reversal. I do not even know how to describe the reversal. I saw a diagram on the Forum where the coordinate axes are referenced to zero or 1 bar and the trends show fans of flies or something else, it means that the price always reaches zero - that is why the person bragged that he caught the bird by the tail. That is my fantasy, and to describe is more difficult than to imagine, but wrote and seems to be something like that. The question is whether something can come out of it. I would also add: at the highest peaks where rebounds usually occur on the average - put a reverse order with a profit on the rebound percentage. I hope it is not too vague - you can also add delayers following the price with some decibels indent to knock them down during sharp price moves, like on the news, for example. Keep it as a secret - what kind of secret is possible here? What has become public domain?)
Here's an idea. I've already put it up for discussion - no one has responded - it's summer - everyone's on canaries or bunk. so, if you do it, or maybe there's already something like this
Here's the idea: introduce price movements in decibels and set zero or zero line on small noises or a flat, when the price goes away, for example by 10%, put an order, depending on the strength of the average squeak calculated from the history, add, say, +51%, set 0 again and go stepwise again. The problem is to catch a candle or the daily trend by at least 30-50%, like this. Well, a reversal: depending on the sine of the angle, it has gone higher than one by 10-20%, which means a reversal. I do not even know how to describe the reversal. I saw a diagram on the Forum where the coordinate axes are referenced to zero or 1 bar and the trends show fans of flies or something else, it means that the price always reaches zero - that is why the person bragged that he caught the bird by the tail. That is my fantasy, and to describe is more difficult than to imagine, but wrote and seems to be something like that. The question is whether something can come out of it. I would also add: at the highest peaks where rebounds usually occur on the average - put a reverse order with a profit on the rebound percentage. I hope it is not too vague - you can also add delayers following the price with some decibels indent to knock them down during sharp price moves, like on the news, for example. Keep it a secret - what kind of secret is possible here? What has become public domain?)
An interesting theory, but no more than a theory ... By the way, several years ago I also made analogies like this with sound... I'll try to tell you my view of the practical side of it...
1. After all, sound is a pure sine wave, with varying amplitude and periods... it's also possible for several different waves to overlap. for a course it's not really that... although it can be assumed that there are so many waves superimposed that the sinusoid looks distorted... (let's assume that it is a sinusoid after all)
2. what it takes to use this theory... you need to remove excess noise and leave a useful signal... you have to put in a lot of formulas and calculate the noise... like, say, a digital beeper does... ...decompose the signal and then reassemble it without noise... honestly, I can't even imagine how to do it without in-depth mathematical analysis (let's say we have these formulas)...
3. digitizing quality... the shorter the wavelength, the fewer values there are to describe it, and the worse the quality of the wave after reconstructing it... Our wave is very short, hence the errors in digitizing will be high (this may turn out to be critical, although you won't know until you try)
4. Forecasting waves in forex does not go as long as desired, i can assume that a few waves may be caught in this way, and then either the amplitude or the period will change ... and besides the wave may still go in a trend ...
5... in the end we'll get an indicator like RSI and it's not necessarily better... and in the worst case it may even lie, because at each of the 4 points we deviate from the ideal theory and pick up errors... Considering how much time, effort and knowledge you have to spend on it all, you really don't want to do it...
6. Alternatively, you could try to calculate it all in Matlab, which would be much easier... I even came across a way to transfer quotes to Matlab somewhere... If you're able to do that in Matlab and show it works, I'm ready to go for it, but I'll have to connect Metatrader to Matlab... for mt4 it's too hard to program it... Or a dll will have to be written...
That is my opinion...
P.s. As it seemed to me decibels are not quite correct to use, they show the level of damping and amplification... but not the end of the wave... I may be wrong...
I can't do without matlab and I wouldn't rely on the indicator but only on the difference of decibels in the form of logarithms or Fermat's theory. There is an article in MT5 containing predictor indicator based on harmonic decomposition Fourier_Extrapolator_of_Price.mq5 - even if I couldn't find it through the search engine. Well, it goes something like this...
I guess this is what you're talking about? https://www.mql5.com/ru/code/130 interesting indicator... have you been following it? how accurately does it predict... I don't think it would be difficult to write an Mt5 Expert Advisor for this indicator... but I have a feeling it will lie... If you don't know what kind of changes you want to make, and if you don't know what kind of changes you want to make, you may use it as a guide ... I'll have a look at it this weekend myself, maybe I should try it...
- votes: 20
- 2010.07.05
- gpwr
- www.mql5.com
That's right, that's the one. I have not had enough patience to catch some probability of prediction, and how to insert this fan into the tester. But as much as I watched it almost always shows against the moment of price movement, but again, depending on the TF and the movement on the news - a candle it can never predict, if only to add a filter. If we want to use its continuation for N bars and pick the best probability with the tester, and if after N bars its continuation comes, it means it has caught the trend, if it is absent, it means the trend has changed. If its continuation goes after N bars then it means that the trend has changed, if it is absent then the trend has changed. So, I even got a headache. It seems to be not for medium-sized brains, and no one is discussing it for free.
The problem is that the traders do not understand what's going on in the market and the trading robot does not do anything ... If you want to trade robotically, you've got to know how to trade robotically - so, you have to work against it.
That's right, that's the one. I have not had enough patience to catch some probability of prediction, and how to insert this fan into the tester. But as much as I watched it almost always shows against the moment of price movement, but again, depending on the TF and the movement on the news - a candle it can never predict, if only to add a filter. If we want to use its continuation for N bars and pick the best probability with the tester, and if after N bars its continuation comes, it means it has caught the trend, if it is absent, it means the trend has changed. If its continuation goes after N bars then it means that the trend has changed, if it is absent then the trend has changed. So, I even got a headache. It seems to be not for the average person, and no one is discussing it for free.
The problem is that the traders do not understand what's going on in the market and the trading robot does not do anything ... If you want to trade robotically, you've got to know how to trade robotically ... So, I have to trade against an indicator - it's what I understand after a pint of beer ... Good luck to everyone and goodbye ...
If you have a good strategy and are willing to share it, I can write an EA. I invite you to discuss either publicly or in private messages.
I would like to rewrite my EA or rewrite my strategy I am ready to pay for your services. 1h-4h-1d EA. Uses mA indicator. I don't know if I can pay for it. I just don't know if I can pay for it, I just don't know if I can pay for it.
Mechanic:
Я могу попробовать. Тоже бесплатно.
If you have any questions, please do not hesitate to contact me.
I downloaded it, I'll look at it during the day, I also wrote an EA based on mA at the time... and it also gave good results but only in one direction... the main trend helps it a lot, in the opposite direction it lags, although I tried to apply it to smaller periods... but it seems to me it doesn't trade in the opposite direction either due to accidental or deliberate error...
Appealing to Mechanic:
If you do it, let me know so I don't have to do the same thing... and the result in the studio:)
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use