Discussion of article "Random Walk and the Trend Indicator" - page 4
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
If you take the market chart by ticks and not by time, there is no flat, there are only fewer ticks per unit of time.
You're wrong. And here's why. Your statement is true only in a logical sense. A trader deals with mathematics, and it says that a tick is the minimum price change, and therefore such a change is the minimum limit of measurement, i.e. measurement error. Therefore, all movements that are within a certain number (depending on the algorithm that operates with this or that way of measurement and calculation (in the simplest case 2 ticks)) of ticks can be considered a flat. This is an eternal philosophical contradiction. However, it tells us that besides logic it is worth remembering the maths so that our statements are consistent.
In addition, some calculation algorithms need to fill empty time intervals, for example, a candlestick indicator. This is the second type of "flat", the existence of which is caused by the very algorithm of using (converting) data.
You're wrong. And here's why. Your statement is true only in the logical sense. A trader also deals with mathematics, and it says that a tick is the minimum price change, and therefore such a change is the minimum limit of measurement, i.e. measurement error. Therefore, all movements that are within a certain number (depending on the algorithm that operates with this or that way of measurement and calculation (in the simplest case 2 ticks)) of ticks can be considered a flat. This is an eternal philosophical contradiction. However, it tells us that besides logic it is worth remembering the maths so that our statements are consistent.
In addition, some calculation algorithms need to fill empty time intervals, for example, a candlestick indicator. This is the second type of "flat", the existence of which is caused by the very algorithm of using (transforming) data.
Unfortunately, you are mistaken, but maybe it is for the best. You are wrong in maths.
1. a tick is a price change, but not necessarily minimal. The price for 1 tick can jump so much (GEP for example).
2. The minimum price change is a point
3. Therefore, the minimum error is equal to point/2, not multiply as you have. The error is equal to half the value of the smallest digit.
4. If we want to operate with non-contradictory concepts and always remember about metathematics, then give a mathematical definition to such a philosophical concept as "trend" and "flat" ...
I dare to assert that a trend is the equation of a straight line y=a*x+b. Based on this mathematical statement, I can say that trend IS always, flat is not, it simply does not exist.
H.Y. for many years I want to see a mathematical definition of flat, something that can be calculated and say that yes, now it is flat, and now it is not...and this formula should work always, on all markets, on any time intervals...(the equation of a straight line = trend, works everywhere and always)...but there is no formula of this mythical flat (maybe I was looking for it badly ? ))). To paraphrase one hero - "Flat is in our heads".
H.y.y So vlad123 is right - there is no spoon ))).
And what would actually change if you did, say, get a definition of flat? Somehow it seems to me that nothing will change? Especially since, as you rightly say, it is a purely philosophical concept.
Moreover, if we develop your thought that a trend is a straight line, although a straight line does not really exist in nature, it turns out that a flat is a straight line going horizontally. Flat is a precursor of a trend directed in the opposite direction, or the first swallow - the first attempt to such a reversal. Flat is a state of equilibrium, but since there is no such thing as a state of equilibrium in nature, a flat is only a "moment"....
so still the author or maybe someone has figured it out and will help
2. How are the values of the trendiness indicator obtained - there are some values there, how are they obtained? I didn't see the formula.
?
Unfortunately you're wrong, but maybe it's for the best. You are wrong about the maths.
1. a tick is a price change, but not necessarily a minimal one. The price for 1 tick can jump so much (GEP for example).
2. The minimum price change is a point
3. Therefore, the minimum error is equal to point/2, not multiply as you have. The error is equal to half the value of the smallest digit.
4. If we want to operate with non-contradictory concepts and always remember about metathematics, then give a mathematical definition to such a philosophical concept as "trend" and "flat" ...
I dare to assert that a trend is the equation of a straight line y=a*x+b. Based on this mathematical statement, I can say that trend IS always, flat is not, it simply does not exist.
H.Y. for many years I want to see a mathematical definition of flat, something that can be calculated and say that yes, now it is flat, and now it is not...and this formula should work always, on all markets, on any time intervals...(the equation of a straight line = trend, works everywhere and always)...but there is no formula of this mythical flat (maybe I was looking for it badly ? ))). To paraphrase one hero - "Flat is in our heads".
Z.y.y So vlad123 is right - there is no spoon ))))
Ok, maybe it's for the best that you are mistaken, and maybe it's unfortunate :).
1) It is clear that a tick can be several points. But a tick is the Minimum price change, because it is defined not only by points, but also by time. And what is the minimum unit of time on a tick chart? 1 tick. I.e. within time, whatever the move is in points, but for 1 tick, it is within the measurement error. This, of course, is not the truth in the last instance, but only my reasoning.
2) The minimum price change, as it seems to me, is a point per tick.
3) If we decompose the error from my previous post into components, it seems to me that it will be expressed in ticks, i.e. in some formula linking price in points and time (number of ticks).
4) So I have already given it above. There are 2 kinds :) 1 - little changes (within the minimum) - cannot be considered as a move (this is for a tick chart), 2 - flat on a discrete time scale.
I dare to say that a trend is an equation of a straight line y=a*x+b. Based on this mathematical statement, I can say that a trend IS always present, flatness does not exist, it simply does not exist.
Ok, then try to estimate the significance level of linear regression at different market sections and for different periods. How to do it is written in statistical textbooks :) Then, it will be interesting to hear the results of your research - whether it is always reliable to represent a section of the price chart by a straight line.
H.Y. for many years I want to see a mathematical definition of flat, that can be calculated, and say that yes now is flat, and now it is not...and this formula should work always, on all markets, on any time intervals... (equation of straight line = trend, works everywhere and always)...but there is no formula of this mythical flat (maybe I was looking for it badly ? ))). To paraphrase one hero - "Flat is in our heads".
You've been looking for a long time. Try to approach the question from the other side and try to define a trend. What is a trend in the market is written in econometrics textbooks. Simply put, it is the presence of deterministic (non-random) components in the price chart. Accordingly, a flat is when the estimation of the significance of such components is small, and the chart is considered more random.
You've been looking for something for a long time. Try to approach the question from the other side, and try to define a trend. What is a trend in the market is written in econometrics textbooks. Simply put, it is the presence of deterministic (non-random) components in the price chart. Accordingly, a flat is when the estimation of the significance of such components is small, and the chart is considered more random.
I am not looking for something that is not there. I just ask this question to those who, in my opinion, can think. I hoped that there would be a constructive discussion, with arguments...You sent me to...read econometrics. Well, I can tell you to chuck it in the bin. And start reading the right books, which can withstand many reprints, for example, this Handbook of Mathematics http://lib.mexmat.ru/books/2964/s10.
And about you introducing new definitions, you don't have to do that. A tick is an event that says there has been a price change ( and nothing else). Price is measured in points, the minimum step is 1 point, time has been measured in seconds, microseconds, it is possible in years...If you do not agree with this, then write to the developers and let them correct these concepts in the documentation and give the correct ones in your opinion, you can refer to econometrics that would be more convincing.
And what would actually change if you did, say, get a definition of flat? Somehow it seems to me that nothing will change? Especially since, as you rightly say, it is a purely philosophical concept.
Moreover, if we develop your thought that a trend is a straight line, although a straight line does not really exist in nature, it turns out that a flat is a straight line going horizontally. Flat is a precursor of a trend directed in the opposite direction, or the first swallow - the first attempt to such a reversal. A flat is a state of equilibrium, but since there is no such thing as a state of equilibrium in nature, a flat is only a "moment"....
not so much, I'm saying there is no flat. It doesn't exist in nature, it's only in our minds. Don't look for a black cat in a dark room, especially if it's not there.
There's no flat, there's only a trend. There is always a trend.
Я не ищю того чего нет.
Alas, the very concept of "trend" exists only in opposition to the concept of "flat", not separately from it. If there is no flat, how do you find a trend - something that does not exist?
I just ask this question to those who, in my opinion, can think.
Sorry, if anything, I answer according to my limited reasoning:) If you want - tell me, I will not.
I hoped that there would be a constructive discussion, with arguments ...
You have provided them - you have offered to test the significance of linear regression. Let me ask you, will there be results, i.e. arguments?
What is wrong in the textbooks on econometrics, in relation to the approaches to trend detection (with arguments, please)?
.
And start reading proper books that have been through many reprints, like this Maths Handbook http://lib.mexmat.ru/books/2964/s10.
Thanks, added it to my library :)
A tick is an event that tells you that there has been a price change ( and nothing else). Price is measured in ppts, minimum step is 1 pip, time all my life has been measured in seconds, microseconds, you can do it in years...
If this is so for you, and other approaches are not interesting to you, it is your right.
If you do not agree with this, then write to the developers and let them correct these concepts in the documentation and give the correct ones in your opinion, you can refer to econometrics, which would be more convincing.
That's what I'm talking about, if they test the "significance of linear regression", you can only test the significance of the coefficients in the equation. According to the definition of mathematicians - A linear regression coefficient is considered significant if its MNC estimate is different from zero....
And how econometricians do it...is a mystery covered with darkness, personally for me. An argument about nothing. I'm not interested. Sorry, I'll try not to answer any more.
...there is no flat - write about it to all developers of trend indicators and people who use them, let them admit that there is no flat. For convincing you can refer to the maths handbook :)