Calculation of the slope angle of the trend line. - page 19
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
calculation of the angle in degrees, and how to get 2 points from the trend is repeatedly cited here
Like a ray of light in a dark kingdom? So out of all the 18 pages of this topic no one knows what the arctangent is and what it is used for?
Bury the unborn baby.
Every idea is subject to scrutiny.
Different paths, different models, different entities. :-)
I'm trying to use the angle of sliding lines (or their derivatives) varying from 0 to 360 degrees. Quite an analogue of the trigonometric phase. I suppose the advantage is a continuous numerical series on one assumed cycle. It is difficult, so far, to judge the results, but as one of the parameters it looks promising.
I'm trying to use the angle of sliding lines (or their derivatives) varying from 0 to 360 degrees. Quite analogous to trigonometric phase. I suppose the advantage is a continuous numerical series on one assumed cycle. It is difficult, so far, to judge the results, but as one of the parameters it looks promising.
The same question to you - what is a "10 degree angle"? I've already given you an example where different people say different things about the same line. If you really like straight angular measure - you should specify scale, and in two directions at once - price per pixel and bars per pixel (but for the angle you need only their ratio, so you can do with one value). Only that way will your angle value be unambiguous.
There, on the previous page - a function that returns the angle taking into account this scaling value is suggested. But what should it be is unclear. Wouldn't it be better to refuse to use it at all?
Forum on trading, automated trading systems and trading strategy testing
Calculation of trendline slope angle.
George Merts, 2017.06.01 05:35
Same question to you - what is a "10 degree angle" ?
If you have a sinusoidal line, whatever the amplitude or period, or however it is stretched along the X or Y axis, you can always unambiguously match the point of the sinusoid to the angle of rotation of the radius - vector or tangent to it.
Forum on trading, automated trading systems and strategy testing
Calculating the slope angle of a trendline.
George Merts, 2017.06.01 05:35
you have to necessarily specify the scale, and in two directions at once - price per pixel and bars per pixel (but, the angle requires only their ratio, so you can get by with one value).
No need. Since the graph is not a perfect sine wave, description of boundary conditions and requirements would be very voluminous. If at all possible. Therefore we are talking about the phase analogue, and it is sufficient that it is calculated using the same algorithm on any chart, of course taking into account the external variables.
Forum on trading, automated trading systems and testing of trading strategies
Calculation of the slope of the trendline.
George Merts, 2017.06.01 05:35
Wouldn't it be more correct to abandon its use altogether ?
If you have a sinusoidal line, whatever its amplitude or period, or however it is stretched along the X or Y axis, you can always unambiguously match the point of the sinusoid to the angle of rotation of the radius-vector or tangent to it.
So, again, you are introducing as much as two additional quantities - amplitude and period.
Not necessarily. Two points are enough to fully define a straight line. Three points are enough to completely define a sinusoid near zero. Accordingly, the three points unambiguously define the phase, while the amplitude and period are likely to be caveated to inflection points.
But again, the graph and the slider are not perfect sinusoids. That's why we are talking about a phase analogue.
I don't know what else to add, including the choice to "see a 10 degree trend" or ....
Not necessarily. Two points are enough to completely define a straight line. Three points are enough to completely determine the sine wave near zero. Accordingly, the phase is unambiguously determined by the three points, while the amplitude and period are likely to be caveated to inflection points.
But again, the graph and the slider are not perfect sinusoids. That's why we are talking about a phase analogue.
I don't know what else to add, including the choice to "see a 10 degree trend" or ...
I don't mind that you can represent the trend with this method. The question is - what are the advantages of it ?
Let's include the DPI of the screen into the trend indicator, let's include the aspect ratio of the chart into the trend indicator - we will get a lot of different variants. But how are they better than what we have?
My opinion is the following.
We need a trend characteristic. It can be expressed using different methods.
The trend, expressed as a purely degree measure of slope - has the greatest comfort for the perception. The trend expressed in points per bar is the most correct and accurate expression of a trend, but it is harder to perceive such a trend.
Accordingly, a simple degree measure will be more suitable for an intuitive visual assessment. A person usually gets used to his or her usual scale and the intuitive comparison starts working.
For Expert Advisors, this approach will not work, the Expert Advisor will not care about the scale, that's why the "trend expression through the price movement per unit time" approach makes much more sense.
All other approaches, such as "degree measure with respect to scale" is extremely inconvenient for visual intuitive trading, and does not make sense for EA trading.
Above - one of the participants suggested to "take into account human psychology", for which it may be useful to measure the trend by a "degree measure with scale". But, again - no examples (for evidence, I'm not even speaking) of usefulness of such an approach - has not been presented yet. We wait...
I don't know if it's possible to determine the trend or the exact slope that doesn't depend on the chart scale.
I have read the thread to the end. You suggest to calculate the angle of slope of the trend through the tangent. it is quite possible, similar to the polar coordinate system. the question is what scale to set? and most likely it will be different in different segments or not? and the most interesting thing is how to determine the pivot point when the slope angle changes and how to understand it in advance? and in general it all reminds of the hannah system. Could you tell me please which direction to take? I've got stuck in my head.
the question is, what scale should I set?
Time is passing, but the tasks remain the same. I have faced the need to calculate the angle of slope of the trend and was about to start doing it on the basis of coordinate points, but by my good fortune I came across this article. I did not take into account that if we are referring to coordinates, the angle will change when zooming.
There is only one solution here, we have to forget about the angle degree in the classical sense. Theoretically, we can construct a triangle where one cathetus is bars and the other cathetus is points, calculate the angle of that triangle. You will get the technical value of the slope angle, it will not change when you change the scale, but you will visually see the angle on the chart quite differently because we have different types of data for the cathetuses.
To my mind, it is better to use the percentage of price change at the moment of time instead of the degrees. After all, it is the numbers by which you build your strategy that are important.