Koen Van de Moortelindependent math & physics tutor
Koen Van de Moortel
Open for challenges: do you have questions about measuring methodology, modeling, regression? Ask me! No cure, no pay.
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
Citations since 2017
11 Research Items
Experimental physicist, looking for professional challenges. 2021: finished writing a book about measuring methodology and regression analysis + software, for students and teachers, in my native language: Dutch. See: book: www.lerenisplezant.be/metenisweten.htm software: www.lerenisplezant.be/fitting.htm If a publisher is interested, I can translate the book in English.
January 1990 - June 2020
- math/physics teacher
FittingKVdm 1.14 has been released! There are many software programs that do regression analysis (curve fitting), BUT only this one makes optimal use of the symmetry whenever dependent and independent variables can be switched! This will definitely improve your scientific modeling or calibration work! And in most cases it will estimate good initial...
(European) shoe sizes of adults versus their height (126 men and 242 women, mostly from Belgium and Holland, Feb. 2023) - an educational example of my regression software "FittingKVdm".
Two methods for finding the "best" curve fitting through a set of data points are evaluated here: "multidirectional" and "ordinary" least squares regression (MDLS and OLS). The same artificial datasets with several amplitudes of generated noise were fed to both algorithms and the results were compared.
This real life example shows why Multidirectional regression is better than the classical method (OLS): it treats the so-called independent and dependent variables equally.
Does a person unconsciously improve his psychomotor while performing a simple task like throwing a pebble? Apparently he/she does, at least with our test person. The hypothesis was tested with 3 different methods: comparing averages, Kendall's tau, and regression.
The so-called 'least squares regression' for mathematical modeling is a widely used technique. It's so common that one might think nothing could be improved to the algorithm anymore. But it can. By minimizing the squares of the differences between measured and predicted values not only in the vertical, but also in the horizontal direction. I call t...
For non-linear regression, often the logarithms of the variables are taken, to reduce the problem to a linear regression. With some examples, I explain you why this is not such a good idea.