Zero sample correlation does not necessarily mean there is no linear relationship - page 10
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
corr(A,B) - calculating the sample Pearson correlation coefficient.
I am very familiar with this Mathcad function. It is calculated as covariance divided by the product of RMS:
Mathcad calculates the sample correlation coefficient exactly as defined. Linear regression has nothing to do with autocorrelation.
Correlation is simply a number that is always obtained by calculating the formula, to everything from -1 to +1. Applying the formula to BP we will always get some correlation value: between any currency pairs, between a currency pair and the movement of Jupiter, between a currency pair and anything else.
.....Count the correlation between price series or between a currency pair and the movement of Jupiter - it makes sense, although it makes little sense. ......
Clarify: so does it make sense or is it faintly obvious????????? Otherwise, it's somehow very vague....
I don't argue that there is no point in calculating the correlation between Granny's BP and Jupiter's motion BP without a corresponding prior BP transformation.
But if you want to look for linear relationships between price BPs via QC, doing so without first logarithming is simply wrong. Without logarithm, the meaning would be the same as with Granny and Jupiter.
Correlation can indeed be calculated for any BP. Calculating the correlation between price series or between a currency pair and the movement of Jupiter, although it makes sense, makes little sense. You have to adequately prepare BP before calculating the correlation to make sense of QC. If the sense is to find a linear relationship between price VRs, then the initial VRs should be prolagarithmic before calculating the QC.
Absolute nonsense.
Correlation can indeed be calculated for any BP. Calculating the correlation between price series or between a currency pair and the movement of Jupiter, although it makes sense, makes little sense. You have to adequately prepare BP before calculating the correlation to make sense of QC. If the sense is to find a linear relationship between price VRs, then the initial VRs should be prolagarithmic before calculating the QC.
Yeah, you have to know how to cook a cat.
The whole topic is a demonstration of a correct, not always correct and generally incorrect understanding of certain mathematical concepts. People, as it seems to them, have managed to understand the formula, and on the basis of the correctness of the calculation by the formula claim the correctness of the result.
There used to be a science called systems analysis. It began with a list of mistakes made by systems engineers. The first error was: solving the wrong problem using the right methods. The complexity of solving a problem is not in the methods of solving it, but in the formulation of the problem. This first error of system analysis is the most widespread in science and engineering. This forum and this thread only confirm this rule. Methods can be learned, nothing complicated and accessible to almost everyone except for some pathologies. But to set the goal correctly is not for everyone and not everyone can be taught. Even if you manage to do it at least once, there is no guarantee that you will be able to do it again.
That is why it is so important to study experience and without reference to experience in full, not in neutered formulas - all formulas, even from Matcad, are an empty sound.
If we are discussing correlation, then without a meaningful justification of the validity of the relationship, and without giving a numerical estimate of the confidence in the result obtained, it is all profanity, which does not allow for practical meaningful results.
Absolute nonsense.
Nonsense is when a correlation (which describes a linear relationship) is applied so that it shows different absolute values for {EURUSD; USDJPY} and {EURUSD; JPYUSD} pairs.
It is nonsense to apply a correlation (which describes a linear relationship) in such a way that it shows different absolute values for {EURUSD; USDJPY} and {EURUSD; JPYUSD} pairs.
It is nonsense to get a correlation value without assessing significance.