Biology, images, analysis, design...
Use/Abuse Principles How To Related
"It has long been an axiom of mine that the little things are infinitely the most important" (Sherlock Holmes)

Search this site



Just a note

The reason it is important to linearize relationships is because it is much easier to quantify and assess linear relationships than non-linear ones. To cope with non-linear relationships requires more complex models, many more assumptions, and far more calculations.

One important reason for this is that regression analysis unavoidably requires some sort of 'fitting', or parameter estimation. For example, unlike a mean (which only estimates one parameter - the population mean) a linear regression estimates two parameters - the slope (b) and the intercept (a). The more parameters you have to estimate, the more complex a relationship you can model, the more information you need to do so - and the more potential there is for spurious relationships, and for unjustifiable assumptions.