Hi Mladen, Mr. Tools and All,

I was searching online for a correlation indicator to compare two currency pairs then I realized that I already had one in my indicators folder since 2017 but forgot about it. I can't even remember where I got it from. I probably just tested it briefly then to see what it was all about. It's called "smCorrelation MultiPairs" and it uses the same formula and gives almost the same results as the correlation tool

Since the correlation results at mataf.net are slightly different, I'm thinking that maybe there is a slight error in the code used to compute Pearson's Correlation Coefficient in this indicator, and if there is an error, then it would be great if one of you can fix it since more accurate correlation results can translate into more profitable trading. I thought that maybe mataf.net used a Length_Calculation different than 50 but I tried many different values but still didn't get the same results they got. Or maybe, this indicator is correct and mataf.net is wrong.

I read that the Pearson formula is only accurate for normal distribution time series whereas the Spearman's correlation formula is accurate for both normal and non-normal distributions making it perfectly suited for analyzing forex pair correlations but since the Spearman calculations take too much CPU resources and the results are not much different than the Pearson correlations then it's more practical to just use the Pearson formula.

However, I got the idea that since the fisher transform can convert a forex time series into an almost normal distribution, couldn't we first apply fisher transform to the prices then use the Pearson correlation formula on that transformed almost normal distribution to get an even more accurate correlation result than using Spearman's formula while using less CPU resources? Mladan, Mr. Tools, Jimmy, etc, what do you think about that? Is that feasible?

Anyway, here is the "smCorrelation MultiPairs" indicator.