Where is the line between fitting and actual patterns? - page 40

 
Figar0:

BackOOS and forwardOOS are the same thing.

Let me explain, if we are not talking about "eternal" patterns that are always in the market, but about coming and going, a pattern appears at some point and disappears at some point. The probability that our Sample period will coincide with the moment the pattern appears is approximately equal to the coincidence of its end with the disappearance of the pattern. By conducting OOS after Sample we increase the chance of not catching a pattern in the market. By conducting it before Sample, we bypass this rake.

In fact, this is a very primitive judgement, a pattern does not appear overnight and does not disappear at one moment. There are always regularities in the market. There are many of them. Perhaps it looks like a pile of sinusoids superimposed on each other with a certain pitch. What is worse, there are a lot of these piles with various periods. Selecting the Sample size we select we select the period of this sine wave. After we run Sample, we should catch the sine wave falling maximally on the Sample time section (this pattern is the easiest to spot). There should be tails in front and behind. What is in the front may be traded, the one behind is suitable for testing.

I use only backOOS for my TS, fOOS can be used for tests and experiments. In principle this approach justifies itself for me.

Z.U. I wasn't able to express my thoughts clearly, but honestly, I tried)


What is it?

Very clearly. At least, you didn't get far from me when it comes to inarticulate phrasing. :)

A wonderful analogy with waves. A crude analogy, but well understood. Rough, because they are not even waves at all, but some kind of curvulines, overlapping each other and sometimes resonating with each other. The main thing is that this perfectly describes the difference between the second type of systems and the first. The latter, due to the limited size of trades, are just catching these waves (periods of waves). And the first ones, it is already clear for what reasons, cannot do it (that is why I say that for them optimisation==fitting).

 
VictorArt:
Sorry, but you missed the point - when combining sets like this it will be impossible to find regularities with known methods - "figures" will be different not only in their size, they will be absolutely incomparable, i.e. you will surely find similar figures, but they will be similar only in their form, but not in their content.

The form unambiguously corresponds to the content, irrespective of size. If this is not the case for you, then I don't know what you are looking for.

The point is to look for and find recurring cause-and-effect relations - regularities. Anything that cannot be directly measured and compared (whether we know how to do it or not yet) is not of interest to a trader whose goal is to make money.

 
joo:

Rough, because they are not even waves at all, but some kind of curvulines, overlapping each other and sometimes resonating with each other.

Well, if they really were proper sinusoids, our task would not be easy) I can not draw, but if you represent these curvulines on the graph, where on the abscissa axis of time, and on the ordinate axis - profit from using curvuline-laws, the short, short-period curvulines will be the highest and steepest waves, long will be much flatter, and lower. About "resonance" you are correct, the coincidence of several waves of comparable period will make these curvilinear waves lower.

It remains to be seen how this can help our research.

-----------

Z.U. Can someone depict the described picture for clarity? Tried it in MSpaint it's creepy.....)

 
Figar0:

Well, if they really were proper sinusoids, our task would not be easy) I can not draw, but if you represent these curvatures on the graph, where on the abscissa axis of time, and on the ordinate axis - profit from using curvature-law, the short, short-period curvatures will be the highest and steepest waves, the long will be much flatter, and lower. About "resonance" you are correct, the coincidence of several waves of comparable period will make these wave-curves lower.

It remains to be seen how this can help our research.

-----------

Z.U. Can someone depict the described picture for clarity? I tried it with MSpaint, it's creepy.....)

And I can draw anything, in this case, do not understand what?

....................................

And I'd also like to see with my own eyes - what are the following statements about?


joo:

The form unambiguously corresponds to the content, regardless of size. If that's not the case for you, then I don't know what you're looking for.

 
Figar0:


The model in this context will not be one and moreover niversalnaya, everyone will take something of their own, which may find a use. And please, do not throw all sorts of bOOS, bOOS2, etc. at our compliant minds. Let's just have Sample and OOS for now).


Forgive me generously, I don't want to bring discord into our fledgling relationship at all,

but how does the preceding correlate with the following?

I don't think so.... ((

Figar0:

BackOOS and forwardOOS are the same thing.

I use only backOOS, you can take fOOS for tests and experiments. In principle this approach justifies itself with me.

Z.U. I was not able to express my thoughts clearly, but honestly, I tried)

 
lasso:

Forgive me generously, I don't want to bring discord into our fledgling relationship at all,

but how does the preceding correlate with the following?

I don't think so.... ((


A specific question was asked there about bOOS. And although we (or I) haven't gotten to that point yet, I thought I'd give my opinion. It's rude to ignore the question of the questioner since we're "frolicking" around here.

Here draw a picture of my 2 posts that were before this one, and Joo's response in between them and see if we get closer to bOOS'es and fOOS'es. I would make a better picture, but I am afraid I can not)

By the way, are the acronyms bOOS and fOOS your brainchild? google doesn't seem to know it....)

 
Figar0:


A specific question was asked there about bOSOOS. And although we (or I) haven't got to that point yet, I felt it necessary to give my opinion. It's rude to ignore the question since we're frolicking around like this.

Here draw a picture of my 2 posts before that, and Joo's answer between them, maybe we'll get closer to bOOS and fOOS. I would make a better picture, but I am afraid I can not)

By the way, are the abbreviations bOOS and fOOS your brainchild? google doesn't know it....)

1) I really don't get the picture. :-(

2) Google only knows what it has indexed. I've been spared that fate so far.... (unlike Reshetov).

.............

PS I think I've got it. Let me try.... In the sound editor to draw........

 
Figar0:

Here draw a picture of my 2 posts before that, and Joo's answer between them, maybe we'll get closer to bOOS and fOOS. I would draw a better picture, but I am afraid I cannot)

Let me ask you as an artist - artist.....

Here you go.




 
lasso:

1) I really don't get it about the drawing. :-(

2) Google only knows what it has indexed. I've been spared that fate so far.... (unlike Reshetov).

.............

PS I think I've got it. Let me try.... In the sound editor to draw........


In the sound editor is a very good decision.

 
lasso:

And I would also very much like to see with my eyes -- what are the following statements about?

Once again:

Lawfulness is a necessary, essential, constantly recurring relationship of phenomena of the real world, determining the stages and forms of formation process, development of phenomena of nature, society and spiritual culture. A distinction is made between general, specific and universal patterns.

A phenomenon is a combination of events.(pr. joo)

I will try to expose the plot with the red maiden gathering plantain more fully. It is convenient to compare a road to a market instrument, while plantain is a concrete name of a regularity. It is clear that many regularities of different origin (natural, physical, artificial ...) may be present at the same time, like a plantain at the roadside, forest plantations, CPM at intersections, power lines, etc. (there are many regularities - general, specific and universal - for a particular market instrument simultaneously). Then by denoting a trade instrument by Q we'll get:

Q=:{A,B,C,D,I,F,J..............},

where A,B,C,D,I,F,J are names of patterns,

i.e. the trading instrument Q has regularities A,B,C,D,I,F,J....

Hence, the relationships of the phenomena (one phenomenon is followed by the corresponding one), will look like this:

_a->a_,

_b->b_,

_c->c_,

..........

_j->j_,

But, the market is a very complex non-stationary thing, in which there are rarely one hundred percent correspondences of phenomena, so, a qualitative characteristic k of a particular pattern would be something like:

k(Q:A)=(_a->a_)*100%

Just the graph of this k will be like a curve, on the abscissa axis is the index number of appearance of the phenomenon, and on the ordinate axis is the percentage expression of conformity of two interrelated phenomena, 0 being totally different phenomena, 100 being totally different phenomena. It is quite clear that patterns that have an average percentage of matching phenomena greater than 50% are of interest to the researcher.



Thus, a process that has no pattern (i.e. any combination will always have a different corresponding combination) is white noise.


PS has highlighted the corresponding concepts in the definition of a pattern with colour.