Optimal moving average - page 8
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
Something substantial is needed, holy water is weak. The whole horror is that we are about to be understood.
Spicy:) (This one's for Lizone).
Grandpa Sigmund's cousin's posts have disappeared, but I'll try to answer.
I don't reject spectral analysis. The matter is different.
My profession is development of special-mathematics for automation of managerial activity. I have learned most of all from making simple decisions: when automating human activity there is no need to try to change it, it is necessary and sufficient to make it more comfortable and less time-consuming. In this sense, an ideal automaton looks like a "sycophant" generating exactly those solutions which will be taken by a human, whose thinking peculiarities the automaton is "adjusted" (adapted) to.
Do you know at least one professional trader who uses the spectrum analysis method?
Because it is pointless to look for process patterns at intervals whose duration is not comparable to the period of the process. Somewhere like this ...
Once again, what prevents you from trying to reconcile these two disparate concepts. you should probably just define the concept of a trend, and focus on sampling rate alignment.
You should probably just define the concept of a trend, and focus on sampling rate equalisation.
There is no such thing as a no-strings-attached price movement. Kickbacks are always and everywhere, to one degree or another. So when saying that the pullbacks have been (or are not) - explain what you mean by rollbacks.
Kickbacks are by how much? For some it is 33%, for others it is 50%, for others 66% or 100%, for the fifth it is some other figure. On that basis - give whatever definition of trend you like.
Some people do not use % calculation of pullbacks at all, but prefer a point calculation.
Also, calculate bounce correctly. You need to understand where to calculate it from. The price does not move by itself, it is being moved. You need to understand why, how and to what extent it is being moved.
For me personally: a 100% pullback (no matter where, when or on what) is a flat, as I said before. A flat is always part of a trend and a trend is always part of a flat, they are components of each other.
I've already said that I don't work with analogue charts like you (and many wanderers here), I only work with digital charts and digital systems based on certain "manipulations" of numerical price series and trades, and so I don't bother with what a trend and a flat is.
Regarding Fourier: I confess I have worked, but gave up this subject, I did not like it, although I have earned about 20% on it in 2 months. It's not my theme.
A pullback is considered a pullback until it exceeds the size of the price movement in points in the opposite direction, relative to which we defer the concept of a pullback.
Moreover, the concept of a price movement and a move are different things. The price may go through, say, the same 100 points differently.
Prices may also be digitized in different ways, or you may follow Sergei's example) https://www.mql5.com/ru/forum/126769/page141#347465.
Fourier has nothing to do with it.
A thought came to mind. We take 1-5-10-20 etc. pips. If the pullback is not more than 100%, the trend continues on the current timeframe. If more than that, then the trend reverses. Moreover, if it exceeds, say, the level of 50 points but does not exceed 100 points, it means a trend reversal and the turning probability increases on a larger timeframe. We should think it over.
A thought came to mind. We take 1-5-10-20 etc. pips. If the pullback is not more than 100%, the trend continues on the current timeframe. If more than that, then the trend reverses. Moreover, if it exceeds, say, the level of 50 points but does not exceed 100 points, then by 50 points the trend reverses and the turning probability increases. We should think it over.
I should also add a factor that such a pullback is not random and I think a study may be done
I only work with numerical charts and numerical systems based on certain "manipulations" of numerical price series and trades, so I don't bother with what's trending and flat.
That's how many people do it - it's the right thing to do. Any chart can be reinterpreted in the context of a trading system ... not just a system but any filter.
And what is your manipulation of the numerical price series? Is it all through the plug?
Freud 19.05.2012 18:13 Comrade Dimitri never replied to this post.
That's the thing, many people talk about trend and flat, making it abstract, so what can you expect from trying to find an abstract concept?
a trend can be either a non-flat movement or the increase in price within a certain interval.
So far, I consider a trend to be a straight line equation - it always is. You may also consider a trend to be a separate function, whose points are equidistant from the price function.
The problem is that the ends of a linear trend line jump and this line exists on a certain interval, and when the interval is changed, the ends of these lines diverge.
From this point of view, I believe that there are no flags, only trends, and that there is an overlap of trends, or one trend overlaying another one.
all these tare squares... This is a search for resonance points. Filters, too, by removing necessary frequencies, reveal these points.
ZS: it's not my author's, it just happened to reveal, that by high-frequency vibrations, which are hinged to low frequencies of lower order, resonate with high-frequency vibrations, which are hinged to low frequencies of higher order.
here is a picture of figure 2 we can build a circuit of oscillations for higher order (extend it as if into the future).