Local Weighting

The size of the local neighborhoods that PROC LOESS uses in performing local fitting is determined by the smoothing parameter value . When , the local neighborhood used at a point contains the fraction of the data points closest to the point . When , all data points are used.

Suppose denotes the number of points in the local neighborhoods and denote the distances in increasing order of the points closest to . The point at distance from is given a weight in the local regression that decreases as the distance from increases. PROC LOESS uses a tricube weight function to define

     

If , then is replaced by in the previous formula, where is the number of predictors in the model.

Finally, note that if a weight variable has been specified using a WEIGHT statement, then is multiplied by the corresponding value of the specified weight variable.