3 ms·
There is nothing wrong with using a non-linear model, though; x^2 or x^3 regressions make sense on many datasets. Non-continuous is also not the perfect termin
by tylerhou 6y ago
There is nothing wrong with using a non-linear model, though; x^2 or x^3 regressions make sense on many datasets.
Non-continuous is also not the perfect terminology, but I argue that it is more precise than non-linear: the chief idea being that the model "changes unpredictably."
- kccqzy 6y agoSure you can argue things however you want, if you also decide to ignore hundreds of years of mathematical terminology.