What is different from other programs?
It works with iteration, which is applicable to any kind of function and often more precise because there is no need to take logarithms of variables or do any unnecessary transformations.
Some measurements are more precise or reliable than others. Most programs just neglect this and produce parameters that are not optimal. Here, the weights of the data points are taken into account, using the standard deviations (confidence intervals/measuring errors) of both x and y variables! They are also used to make a realistic estimation of the parameter confidence intervals, taking the error propagation through the model function into account!
Most programs just produce a ton of numbers and graphs, and you have no idea how they were produced and how reliable they are. With this program you can visually follow the iteration proces and test what happens if you change the parameters, and that can give you a lot of insight in the stability and the suitability of the model! This also makes it interesting to show in a classroom!
Model functions in this version: constant, linear, quadratic, cubic, exponential+baseline, 1-exponential, Gauss, double Gauss, power, logistic, double logistic (peak), sine, double sine, damped sine, sine with harmonics, parallax, rational (Michaelis-Menten) and another one, refractive index.
Being an experimental physicist, I have personally tested every model with real world data, from physics, biology, agriculture, meteorology, psychology, demography, geography, sports,... These examples are included for educational purposes. You can find them in the 'manual' too.
I will add more model functions and examples according to your needs! And since this is version 1, you will get some free updates for a while!
Improved regression algorithm:
Normally, the model parameters are optimized by minimizing the sum of the squares of the predicted minus the observed y values. This is called 'Ordinary Least Squares fitting'. This works reasonably well in case the data don't have much noise.
Unique in this program: in the case of invertible model functions, the fitting can be done in x and y directions, which gives a dramatical improvement!
I call this 'multidirectional regression' (not to be confused with 'total least squares' or 'Deming/orthogonal regression').
Especially for power function models with noisy data (e.g. biometric data) my program produces way better results than other well known software (e.g. Wolfram, GraphPad Prism, GeoGebra, Graphmatica, TI-84, NCSS,...)! Don't believe me, just try it!
Remark: Stock market gurus or other economic number crunchers will not have benefits from this, since they only study trends, not causal relationships.
Ordinary and multidirectional fitting applied to mass vs height data of adult men with fat% between 11.6 and 13.8, showing that the BMI should be closer to m/h³ than to m/h².
You don't know what to do with your data? Trouble finding a good experiment setup, a measurement strategy, calibration? Confused about all the methods that exist? Questions about the most efficient usage of this program?
I can assist you personally, in simple words, because I hate unnecessary complicated technical mumbo jumbo! In spite of my long experience in math tutoring, of course I don't have all the answers, but asking is free! No cure, no pay!
The first version of this software program is released the 10th of April, 2021. The current version 1.2.2 (6 May 2022) has some extra models and examples.
Requirements: pc with Windows 7 or later/Mac with Windows emulator; recommended screen resolution: 1920x1080 (full HD).
Author: Koen Van de moortel - firstname.lastname@example.org - +32 9 2277036 or +32 47 7368526 - Presentation on Youtube
Support: Fire your questions & requests! I'm not sure if I can answer all of them, but I'll try! - User discussion forum