What is the most efficient way to obtain a multi-variate polynomial equation from a large data set?
1 Ansicht (letzte 30 Tage)
Ältere Kommentare anzeigen
My problem is as follows. I have 10 independent variables that I have permuted to arrive at about 300 different tests. Now, each of these tests give me a two-dimensional trajectory data of a partricle that is easily attained by a 6th-degree polynomical equation that relates the y-distance to x-distance. But my goal, however, is to take these 300 different 6th-degree polynomial equations (please see the plots in the image below) and incorporate with them, the 10 independent variables that I permuted to obtain these 300 different trajectories / equations. Or in other words, I would like to obtain a single grand equation that has a total of 12 variables (the 10 independent controlled ones, and the two dependent ones) that can describe (with an undertandable margin of error) all of the 300 tests. My question is two-fold: i) mathematically, how would I go about doing this? and second ii) what module in MATLAB will allow me to do this most efficiently?. I have some experience with systems identification analysis (linear time-invariant) and very primitive single input regression analysis. However, I seem to be beyond my depth with this current conundrum. My initial hunch seems to direct me towards machine learning; however, I am not sure. Thank you for your time.
2 Kommentare
Jon
am 21 Feb. 2020
So just to clarify, is y a function of x and the 10 independent variables, lets call them v1,v2,...v10? Does each colored curve in your plot correspond to y vs x for a particular set of values of v1,v2,...v10?
Antworten (0)
Siehe auch
Kategorien
Mehr zu Descriptive Statistics finden Sie in Help Center und File Exchange
Produkte
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!