Math Is Fun Forum

  Discussion about math, puzzles, games and fun.   Useful symbols: ÷ × ½ √ ∞ ≠ ≤ ≥ ≈ ⇒ ± ∈ Δ θ ∴ ∑ ∫ • π ƒ -¹ ² ³ °

You are not logged in.

#1 2021-05-26 12:36:03

Registered: 2012-06-14
Posts: 70

New Single Variable and Mulitvariate Data Interpolation Method (Easy)

Just published this video.
Easiest (Single Variable and Mulitvariate) Data Interpolation Method ever made? (May 2021)

I'm not claiming it to be good or useful.  It is what it is.  If it (or the ideas in it) are not new, I would love for someone to point me to older existing sources!

Video Description:
Formed from the simplest of ideas comes potentially the easiest Multivariate data interpolation method ever published!

In this video I present and derive from scratch -- showing several discrete examples and graphs -- a new method for constructing a continuous function to fit ANY data with any number of independent variables!

Unlike other interpolation methods, there is very little difference in the difficulty between the univariate and multivariate case.

The "format"/form/formula for constructing the curve is simply just a skeleton of the independent and independent variable values, accompanied by a predefined SPECIAL function.

NO COMPUTATION is required to actually write out the equation of the curve and very little simplification can be done to the expression once you write it out.  In addition, the resulting continuous function/curve can be more easily computed than polynomial interpolation, especially when doing the calculations by hand and for larger data sets.

I understand that this method will not be useful for fitting many types of data, but I wanted to publish this idea just in case someone sees more in its potential than I do.  Real world applications should not be hard to find!

Please comment sharing your thoughts, and if you would like to contact me, just comment!


Board footer

Powered by FluxBB