How do you properly compare two non-overlapping rows? - page 3
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
Can
Reminds me of a joke:
- Navigator, instruments!
- Fourteen.
- Fourteen what?
- What the hell are the instruments!?
Thank you, I see.
title or link if you don't mind?
standardisation of time series
Okay, read it.
I've done it, I've tried it.
the fiasco comes when the maximum or minimum begins to shift
So the problem with econometric, statistical methods and neural networks is just that.
they all work.
but
on a flat
If you move the price up or down sharply or significantly, it's an embarrassment.Okay, read it.
Did that, tried that.
the fiasco comes when the maximum or minimum begins to shift
So the problem with econometric, statistical methods and neural networks is just that.
they all work.
but
on a flat
If you move the price up or down sharply or significantly, it's an embarrassment.No, I haven't tried it.
No, that's not the problem.
No, I haven't.
no, that is not the problem
There are two non-overlapping rows that are on 'different levels' (like in the picture above).
How can they be 'combined' so that they are side by side and overlap?
You can calculate an average in each row, then row_1 = value_1/mean_1, etc. But is this the right way to do it? Does the sample size affect the adequacy of the results... or should it be done differently? Or through normalisation of Max and Min ? Again sampling period? Actually what is the right way?
I think you know what I mean...
This is done simply.
1. You select the unit of time (conversion interval).
2. Find the arithmetic mean of the two graphs
3. Subtract one value from the other and there is no need to combine anything
This is done simply.
1. You select a time unit (conversion interval)
2. Find the arithmetic mean of the two graphs
3. Subtract one value from the other and you do not need to combine anything
Actually there are infinitely many comparison criteria and methods, and they are all equal in some sense, since the stop-starter didn't quite describe the purpose of the comparison.
I already mentioned MNC - usually it is enough for most problems of series alignment, you can also remember MGC and other methods, it all depends on the task, purpose.
That is, there is no single correct way.
In fact, there are infinitely many comparison criteria and methods and all of them are equal in some sense, as the stop-starter did not quite describe the purpose of comparison.
I already mentioned MNC - usually it is enough for most problems of series alignment, you can also remember MGC and other methods, it all depends on the task and goal.
That is, there is no single correct way.
Here very few people set a problem by final goal (probably, they are afraid that they will stealthe "Grail").
In fact, there are infinitely many comparison criteria and methods and all of them are equal in some sense, as the stop-starter did not quite describe the purpose of comparison.
I already mentioned MNC - usually it is enough for most problems of series alignment, you can also remember MGC and other methods, it all depends on the task and goal.
That is, there is no single correct method.
MNC is usually used for linear dependencies. The most common case is EOPS, which is absorbing and a case of OLS.
OLS - usually used for linear dependencies. The most common case is PNB, absorbing and the case of OLS.
Top of the line...