Chapter Contents |
Previous |
Next |
The KDE Procedure |
To bin a set of weighted univariate data X1,X2, ... ,Xn to a grid x1, x2, ... ,xg, simply assign each sample Xi, together with its weight Wi, to the nearest grid point xj (also called the bin center). When binning is completed, each grid point xi has an associated number ci, which is the sum total of all the weights that correspond to sample points that have been assigned to xi. These cis are known as the "bin counts."
This procedure replaces the data (Xi,Wi), i = 1,2, ... ,n with the smaller set (xi,ci), i = 1,2, ... ,g, and the estimation is carried out with this new data. This is so-called "simple binning," as versus the finer "linear binning" described in Wand (1993). PROC KDE uses simple binning for the sake of faster and easier implementation. Also, it is assumed that the bin centers x1,x2, ... ,xg are equally spaced and in increasing order. In addition, assume for notational convenience that and, therefore, .
If you replace the data (Xi,Wi), i = 1,2, ... ,n with (xi,ci), i = 1,2, ... ,g, the weighted estimator then becomes
The same idea of binning works similarly with bivariate data, where you estimate over the grid matrix grid = gridX×gridY as follows.
Chapter Contents |
Previous |
Next |
Top |
Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.