Discussion of article "R-squared as an estimation of quality of the strategy balance curve" - page 7

 
R2 and shows the slope angle + minimum variance if chO
 
Vasiliy Sokolov:

Okay. Just first calculate the angle of inclination formed by the next two points:

PointValueTime
№112018.03.21
№2103482018.08.13

You decide on a reference space. For example 800na600 we have a drawing, from left to right it is completely filled with a chart, with the top and bottom we have the extremes of the chart. The last point is 50% of the upper extremum, so 400 points. So we calculate it somehow, or what is the difficulty here?

 
Aleksey Vyazmikin:

You define a space reference. For example, we have an 800 by 600 drawing, from left to right it is completely filled with the chart, and the top and bottom are the extrema of the chart. The last point is 50% of the upper extremum, so 400 points. And then we calculate it somehow, or what is the difficulty?

If you need a "space benchmark", please: 800 by 600 pixels. Calculate the angle on it for the two previous points.

 
Vasiliy Sokolov:

If you need a "space reference", please: 800 by 600 pixels. Calculate the angle on it for the previous two points.

You don't seem to be reading what I'm writing. What are these points, where are they located?

 
Aleksey Vyazmikin:

You don't seem to be reading what I'm writing. What are these dots, where are they?

The dots are on the graph.

Can't you plot it? How come? You're weak, aren't you?

 
Vasiliy Sokolov:

The points are on the graph.

You can't plot it? How come? Weak, huh?

Hmmm.... Did you see the picture where the man suggested counting the angle? Doesn't it seem to you that it is defined where the points are?

 
Aleksey Vyazmikin:

You don't seem to read what I write.

... Wrong again, I have read all this nonsense of yours carefully, and specifically offered you a problem that is impossible to solve. Is it still not clear that there are no "extrema", "50%", pixels and other nonsense that you broadcast here? A line connecting two points does not have an "angle". Because the presented graph is not a segment on a two-dimensional Euclidean plane.

 
Vasiliy Sokolov:

... Again you are wrong, I have read all this nonsense of yours carefully, and specifically offered you a problem that is impossible to solve. Is it still not clear that there are no "extrema", "50%", pixels and other nonsense you broadcast here? A line connecting two points does not have an "angle". Because the graph presented is not a line segment on a two-dimensional Euclidean plane.

So you are being asked to do a transformation to a two-dimensional surface, so how can it not be understood? I understand when the problem arises with the chart, because there is no reference space - the aspect ratio of the monitor and its extension, but here....

Let's get to the simple, fix the window of the strategy optimiser, save the drawing of the chart after optimisation in the strategy tester, do the last step 5 times, combine the drawings in photoshop one after another and print them, and now measure their angle, what is the problem?

 
Aleksey Vyazmikin:

So you are asked to do the conversion to a 2D surface, how can it be unclear? I understand when the problem arises with a chart, as there is no space reference - the monitor aspect ratio and its extension, but here...

Let's keep it simple, fix the window of the strategy optimiser, save the drawing of the chart after optimisation in the strategy tester, do the last step 5 times, combine the drawings in photoshop one after another and print them, and now measure their angle, what is the problem?

I bought a kilo of pasta. Question: What is their volume?

 
Vitaly Muzichenko:

I bought a kilo of pasta. Question: What's their volume?

That's an interesting question. It is not difficult to calculate the approximate volume of pasta as a product, it will be more difficult if it is horns or pasta with holes and you need to calculate the volume of the material from which the pasta is made. In the first case for an approximate calculation is enough to take a container in the form of a pencil case and put the pasta there, distribute them evenly and make measurements, in the second case will have to deform the pasta, but it will be enough not a large percentage, if they are the same shape and size .... You can of course also use density, but it's more complicated....