Looking for patterns - page 149
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
The share price may have fallen because of more advanced competitors, but that did not make the company bankrupt, it just made less money.
For some reason I remembered geometry on the ellipse, I remember my teacher taught me how to draw an ellipse using a string, and I tied two nails with a loose string and drew the ellipse using the string.
In figure a triangle is red, so its two folded sides are always equal, in any point of an ellipse.
What I mean is remember there was a topic not long ago, when some TC here what formula he showed, I forget this topic the author showed tricks. So this focal distance is also called focus)))). Maybe he took the formula from there....
For some reason I remembered geometry on the ellipse, I remember my teacher taught me how to draw an ellipse using a string, and I tied two nails with a loose string and drew the ellipse using the string.
In figure a triangle is red, so its two folded sides are always equal, in any point of an ellipse.
What I mean is remember there was a topic not long ago, when some TC here what formula he showed, I forget this topic the author showed tricks. So this focal distance is also called focus)))). Maybe he took the formula from there....
No matter how you twist it in this case the focus, the resulting sample will be perfectly normally distributed
;))
There areregularities in the market:)
For I was selling 2 of these myself in 2014. Just watch the market...
In general, as in any business, trading is 90% luck because those reversals that were in 2008 are not like 13-14, but what about now :) For example, where I trade on vanilla options and I'll tell you that those who with November's contracts last year made at least 10 000% on oil at 25, and now some "analysts" are pushing 5 dollars a barrel, you know how many openers are sitting who wait and pray for 20 for BRENT, I didn't make it so graciously, unfortunately))) And if you remember 13 - I started my capital straight out of the classics from junk fund papers. So a lot of people are drinking champagne now.
And if you remember the start of '14-when the ruble collapsed and the tenge did not.... How much could they do in a day? ))) And there are such regularities even now only not so graceful :).
And gold, my favourite gold, because many people expect prices in the future to go over 2000 :) but no, gold has fooled everybody again like it did in 2013 :)
As for oil, it is a dark horse and no one has managed to tame it... For it is the only instrument which in my personal opinion is not amenable to any analysis...
Look for money flows - for the pattern and its secret is where and where the money flows to and how the crowd behaves ... Look at the securities on NASDAQ for an example and over a 3 year period everything is clear :)
I have another question: how is it that mosbirzhi shares fell in value in 2018 but the company made money and paid dividends?
The earnings of companies have nothing to do with the share price
No matter how you twist the focus in this case, the resulting sample will be perfectly normally distributed
;))
I've thought about it too, there's something magical about it, I don't know how to adapt it yet...
It's getting kind of quiet. Maybe we should continue with the purity of the experiment.
In general, the title of the topic didn't correspond much to the content here. Searching is searching, and checking the suggested ones is another. But thanks all the same. In general, I propose to discuss methods of finding and testing patterns and evaluating their usefulness under the terms of a simplified problem.
Initially the price is a price series with two-way quote by time, or tick numbers. The task is to make trades (on a time interval find the price differences considering the sign up-down) I.e. at the moment of a new tick receipt fix the beginning of the period or the end, with the rule of loss of a two-way quote difference of a spread. There are three parameters on the input. We know the history, the future tick is not. There is no limit on the number of trades. Target, the maximum amount of differences.
Additional input parameters also include volumes, but they are poorly considered in ticks and in their pure form complicate the task, therefore let's simplify and assume that volumes in ticks are equal.
We have all sorts of options to reduce the amount of information. We have thinning, classical candlesticks, Range, Renko, Boxplot and all kinds of averaging by means of which we build more understandable charts and tables and look for regularities)))
We also understand that the price series is a function not of time, but of many factors not considered by us, both global and small.
Time has no effect on price.
We also understand that small factors (actions of some traders or some other ones, which are small and there are a lot of them) have a synergy and their actions may not add up mathematically at all.
From the obvious global patterns, it is of course temporal, year, quarter, time of year, month, week, day, and among the day to see when it is better or worse to make trades.
Searching is usually performed by specifying conditions of deals and considering profit and equity, though it may be done in some other way.
Regularities based on recurring events have also been found. They are usually evaluated the same way.
Generally, due to understanding that the price is not random but consists of many factors, we can describe price movements in terms of behavior. In general, we should introduce the concept of price behavior. We know almost exactly that the price in normal conditions between the ticks has some average difference, and we can calculate the difference with one-millionth or billionth probability on the history, but which were in real conditions and make the corresponding behavioral ranking of the difference between the ticks. We don't take gaps, that's a separate story. Simplify. In general we can do the same with respect to thinned data. In the same way we can estimate other averaged parameters.
Studying of spread and swap behavior in time and their influence on the price behavior is a separate topic, but it will also give some probabilistic result. Although it is still a dark forest for me.
In general, the task is to choose the most complete and significant parameters that describe the price behavior, which can be recorded and regularities can be found. Correlations and Fourier will of course help, as well as statanalysis with NM NM and GA.
I have found this thing, not a pattern yet, but a favourite price behaviour. When volatility subsides, e.g. in the evening, price approaches a round value and starts crossing it back and forth. These are not isolated examples, but a lot.
The red straight line here is a round level multiple of 25 pips, and the yellow lines are a multiple of 5 pips (four digits). And the price spread from the circular level is up to 10 pips each way.
I have found this thing, not a pattern yet, but a favourite price behaviour. When the volatility subsides, e.g. in the evening, the price approaches a round value and starts crossing it back and forth. These are not isolated examples, but a lot of them.
And then what? ....
It jumps, and then ..... goes where?
This, in my opinion, is not a pattern, but anarchy (the one that is the mother of order ))))))))))))))
Wait, Sergei, it's too early to jump, wait!
Just kidding of course, it's not a call to action. It may be trivial somewhere, but sometimes you need to emphasise the nuances in order to think.
For example: to determine with what probability the price finishes its volatile daily move to hang around a level, and with what probability it manoeuvres between levels. What is the swing at those levels (average, median). The duration of this action.