Adaptive Regression

Multivariate adaptive regression splines is a nonparametric technique that combines regression splines and model selection methods. It extends linear models to analyze nonlinear dependencies and to produce parsimonious models that do not overfit the data and thus have good predictive power.

Take a look at the historical Dow Jones Indices below (you can download the data from here).

Let's analyze this dataset to demonstrate the capabilities of PROC ADAPTIVEREG.

Above graph shows 'knots' which are basically end points of periods which can be approximated linearly each. This has a consequence for time-series data; these knot-points can indicate some major effect taking place around that knot.