Dual Purpose Pine Based Correlation

22 mars 2020

This is my "Pine-based" correlation() function written in raw Pine Script. Other names applied to it are "Pearson Correlation", "Pearson's r", and one I can never remember being "Pearson Product-Moment Correlation Coefficient(PPMCC)". There is two basic ways to utilize this script. One is checking correlation with another asset such as the S&P 500 (provided as a default). The second is using it as a handy independent indicator correlated to time using Pine's bar_index variable. Also, this is in fact two separate correlation indicators with independent period adjustments, so I guess you could say this indicator has a dual purpose split personality. My intention was to take standard old correlation and apply a novel approach to it, and see what happens. Either way you use it, I hope you may find it most helpful enough to add to your daily TV tool belt.

You will notice I used the Pine built-in correlation() in combination with my custom function, so it shows they are precisely equal, even when the first two correlation() parameters are reversed on purpose or by accident. Additionally, there's an interesting technique to provide a visually appealing line with two overlapping plot()s combined together. I'm sure many members may find that plotting tactic useful when a bird's nest of plotting is occurring on the overlay pane in some scenarios. One more thing about correlation is it's always confined to +/-1.0 irregardless of time intervals or the asset(s) it is applied to, making it a unique oscillator.

As always, I have included advanced Pine programming techniques that conform to proper "Pine Etiquette". For those of you who are newcomers to Pine Script, this code release may also help you comprehend the "Power of Pine" by employing advanced programming techniques in Pine exhibiting code utilization in a most effective manner. One of the many tricks I applied here was providing floating point number safeties for _correlation(). While it cannot effectively use a floating point number, it won't error out in the event this should occur especially when applying "dominant cycle periods" to it, IF you might attempt this.

NOTICE: You may have observed there is a sqrt() custom function and you may be thinking... "Did he just sick and twistedly overwrite the Pine built-in sqrt() function?" The answer is... YES, I am and yes I did! One thing I noticed, is that it does provide slightly higher accuracy precision decimal places compared to the Pine built-in sqrt(). Be forewarned, "MY" sqrt() is technically speaking slower than snail snot compared to the native Pine sqrt(), so I wouldn't advise actually using it religiously in other scripts as a daily habit. It is seemingly doing quite well in combination with these simple calculations without being "sluggish". Lastly, of course you may always just delete the custom sqrt() function, [Add to Chart] via Pine Editor, and then the script will still operate flawlessly, yet more efficiently.

Features List Includes:
Dark Background - Easily disabled in indicator Settings->Style for "Light" charts or with Pine commenting
AND much, much more... You have the source!

The comments section below is solely just for commenting and other remarks, ideas, compliments, etc... regarding only this indicator, not others. When available time provides itself, I will consider your inquiries, thoughts, and concepts presented below in the comments section, should you have any questions or comments regarding this indicator. When my indicators achieve more prevalent use by TV members, I may implement more ideas when they present themselves as worthy additions. As always, "Like" it if you simply just like it with a proper thumbs up, and also return to my scripts list occasionally for additional postings. Have a profitable future everyone!

11 avr. 2020

Notes de version

Updated sqrt() function. My "intention" of overwriting sqrt() was not to effectively SURPASS the Pine built-in sqrt() function. Please understand this. It was only to provide one of a few computational methods to calculate the estimated square root of a number in native Pine without having intended detriment to this specific correlation algorithm. While this correlation algorithm is not, to my knowledge, prone to encountering negative numbers within sqrt() as it's used, other scripts that may utilize this function may need to be checked for a negative number prior to utilization within sqrt(). I still don't recommend use of this sqrt() function unless you only wish "toy" around with the fascinations of Pine. It's actually a good idea to check for negative numbers on a regular basis when the where abouts of a sqrt() parameter would be potentially unknown. This is a similar situation to checking for division by a potential zero, prior to actually dividing by zero, which is considered an illegal attempt computationally. The square root of a negative number would actually be a complex imaginary number. Pine is ONLY built for real number applications. If this sqrt() function does encounter a negative number, 'na' will be returned now.