LOESS Procedure

The LOESS procedure implements a nonparametric method for estimating regression surfaces. The LOESS procedure allows great flexibility because no assumptions about the parametric form of the regression surface are needed. The LOESS procedure is suitable when there are outliers in the data and a robust fitting method is necessary. PROC LOESS fits nonparametric models, supports the use of multidimensional data, supports both direct and interpolated fitting using kd trees, and performs statistical inference.

Chapter Contents
Previous
Next
Top of Page