Searching for market patterns - page 106
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
AlexeyFX:
I propose to divide the logical and random components of price at the stage of calculation of indicator lines. The random price fluctuations should not affect the regular indicator line, while the regular price movements should not affect the random line. A full view of the price should be maintained, so random fluctuations should not be discarded, as is common practice.
It looks like you can't do without a picture.
The green line is the close price chart. Blue - regular component, unmistakably extrapolated in this case by approximately 32 bars. Red - random price oscillations, hangs around zero, theoretically with unlimited amplitude, but practice shows that it is possible to draw lines that intersect very seldom. Blue line + red line = green at any point. At the same time the blue one can be seen 32 bars ahead. Don't you think you would have to try very hard to lose on this? It is the simplest way of identifying and using patterns, so primitive that it is not a shame to show it to everyone.
I don't see anything useful in separating downward movement and upward movement. It's just another graphic representation of the same thing, while making it more complicated and less understandable, at least to me. You can change the coordinate system and display the price as a spiral or as a completely black spherical thing in a vacuum, what will it change? It is still unclear what to do with it. I also do not see anything good in excluding time from the chart. When I open a trade, I somehow do not really care if I close it in an hour or a year. So the timing has to be at least that's why. There are other reasons too.
And how were built each of these lines . And another vertical line - a reference point should probably be reestablished at each new bar, which will also introduce changes. and where are the values -4.98 and 1.93
To begin with, the reference point is placed anywhere and the close[to] price of the bar where it is located is determined. All prices are calculated using the formula (close[i]/close[to]-1.0)*100.0. Using these values, the green line is drawn, which exactly repeats the price chart, but is expressed as a percentage value relative to the reference point. The values -4.98 and 1.93 exactly show those percentages. Further values of the green line are passed through the lowpass filter, for example МА, and a blue line is drawn. The red line is green minus blue.
There is no point in rearranging the reference point on each bar. Its position affects only the position of the blue and green lines relative to zero, their shape does not change. All other lines do not react on change of the reference point.
I want to call red line a high-pass filter, but unfortunately it is not so, it passes something that shouldn't be passed by the LPF. There are no perfect filters yet, they all introduce amplitude and phase distortion. Phase is much scarier than amplitude, so I'm trying to remove it as much as possible.
This is what a "normal" filter looks like:
It might be possible to trade somehow, but I don't really feel like it.
And here is an "unusual" filter:
You can also do this:
I don't know how you can lose here, although not all the bad effects have been eliminated.
To begin with, a reference point is placed in any place and the close[to] price of the bar where it is located is determined. All prices are recalculated by formula (close[i]/close[to]-1.0)*100.0. Using these values, the green line is drawn, which exactly repeats the price chart, but is expressed as a percentage value relative to the reference point. The values -4.98 and 1.93 exactly show those percentages. Further values of the green line are passed through the lowpass filter, for example МА, and a blue line is drawn. The red line is green minus blue.
There is no point in rearranging the reference point on each bar. Its position affects only the position of the blue and green lines relative to zero, their shape does not change. All other lines do not react on change of the reference point.
I want to call red line a high-pass filter, but unfortunately it is not so, it passes something that shouldn't be passed by the LPF. There are no perfect filters yet, they all introduce amplitude and phase distortion. Phase is much scarier than amplitude, so I'm trying to remove it as much as possible.
This is what a "normal" filter looks like:
It might be possible to trade somehow, but I don't really feel like it.
And here is an "unusual" filter:
You can also do this:
I don't know how you can lose here, although the bad effects are not all eliminated.
To begin with, a reference point is placed in any place and the close[to] price of the bar where it is located is determined. All prices are recalculated by formula (close[i]/close[to]-1.0)*100.0. Using these values, the green line is drawn, which exactly repeats the price chart, but is expressed as a percentage value relative to the reference point. The values -4.98 and 1.93 exactly show those percentages. Further values of the green line are passed through the lowpass filter, for example МА, and a blue line is drawn. The red line is green minus blue.
There is no point in rearranging the reference point on each bar. Its position affects only the position of the blue and green lines relative to zero, their shape does not change. All other lines do not react on change of the reference point.
I want to call red line a high-pass filter, but unfortunately it is not so, it passes something that shouldn't be passed by the LPF. There are no perfect filters yet, they all introduce amplitude and phase distortion. Phase is much scarier than amplitude, so I'm trying to remove it as much as possible.
This is what a "normal" filter looks like:
It might be possible to trade somehow, but I don't really feel like it.
And here is an "unusual" filter:
You can also do this:
I don't know how you can lose here, although the bad effects are not all eliminated.
I have a question - all 17 pairs used by you, they are used only to build indexes, and then you get anticipation, and did you get the same anticipation using only data of the pair?
These are pictures from a test indicator, it only uses data of 1 pair. The real work will be done with 2 separate currencies. Since with me the result of dividing the indices is always equal to the price of the pair, there is no difference and it will not give any additional anticipation, only the possibility to select the most promising and safest pairs.
By the way, by anticipation I mean only the possibility of calculating some regular component of the price due to its past values. I have never tried to get some signals from other pairs or indices before the pair's reaction occurs. I do not believe in it, and I do not need it.
What the fuck are the laws, compare the charts of different DCs, one is like this and the other is like that.
They do whatever they want with you.
SZZY: In this case, it is not the market that determines the price but the one who manages it.
ZZZY: Do you really ever compare the charts of different brokerage companies, even different banks?
These are pictures from a test indicator, it only uses data from 1 pair. The real work will be on 2 separate currencies. Since the result of division of indexes is always equal to price of a pair, there is no difference and it will not give any additional anticipation, only a possibility to select the most promising and safest pairs.
By the way, by anticipation I mean only possibility to calculate some regular component of price, conditioned by its past values
.I have never tried to get some signals based on data of other pairs or indices before the pair's reaction occurs. I don't believe in it and I don't see the need for it.
trol222:
It seems to me that I've got it in my head that you cannot get anticipation without multicurrency analysis and the more pairs for each currency the better (as ardent supporters of spectrum analysis say , and all those who understand it - they say dynamic weights are needed to calculate indexes, and the anticipation itself is obtained from the index analysis - so there goes the paradox without indexes (and without the set of pairs) - no anticipation). I am listening to you with bated breath and I am puzzled. On the one hand with nice pictures and good positions, the anticipation is getting from the same pair, and the indices are needed for more profitable moments for different pairs, but on the other hand they are ..... I am confused - halp me....
I've heard about dynamic weights and anticipation from the indices only from one person. I think he understands something else by anticipation, but I cannot understand what it is. And I doubt he will tell me his secrets. And trying to repeat someone else's work without understanding how it works takes forever. I think it is better to formulate your own task clearly: what exactly do you want to get.
1In this light, it is not at all logical, what is the true price and why the price of the pair must inevitably fall back from it (based on data from one pair).
2-The whole beauty of what you see in your pictures is that the lines continued into the future make it known that a movement from a smaller inflection to a bigger one has begun (without changing the phase of the smaller inflection to the beginning of the larger one) ..... in a pencil, if a small bend appears, it is not certain that a larger one will follow because small bends may not change phase.
3- If someone who knows very well about digital filters probably already understands the principle of your lines, for me it is still at the comprehension stage, I think (I may be wrong) that any pattern, if it is present (and it is present) can be seen by elementary laws of mathematics - +-/* (though messy at first stage) (mathematical complications I think are good - because they can only improve the appearance of this pattern more clearly). All the same, I think we are missing some other constituent link - people do not apply the same laws to what they should, so they do not work in everything that is in the public domain ....... Poor Fourier has suffered so much slander on the forum, because it is not applied there.... .
4 - I'm ashamed to ask)) would you mind making a picture of your preemptive error with arbitrary 2 pairs.
I attached an excel file with 2 excel files of different pairs at 4000 bars the first column - the relative prices, the 2nd - their sum - the trend. If it is not difficult for you. Thanks anyway.
To begin with, a reference point is placed in any place and the close[to] price of the bar where it is located is determined. All prices are recalculated by formula (close[i]/close[to]-1.0)*100.0. Using these values, the green line is drawn, which exactly repeats the price chart, but is expressed as a percentage value relative to the reference point. The values -4.98 and 1.93 exactly show those percentages. Further values of the green line are passed through the lowpass filter, for example МА, and a blue line is drawn. The red line is green minus blue.
There is no point in rearranging the reference point on each bar. Its position affects only the position of the blue and green lines relative to zero, their shape does not change. All other lines do not react on change of the reference point.
I want to call red line a high-pass filter, but unfortunately it is not so, it passes something that shouldn't be passed by the LPF. There are no perfect filters yet, they all introduce amplitude and phase distortion. Phase is much scarier than amplitude, so I'm trying to remove it as much as possible.
This is what a "normal" filter looks like:
It might be possible to trade somehow, but I don't really feel like it.
And here is an "unusual" filter:
You can also do this:
I don't know how you can lose here, although the bad effects are not all eliminated.