| Unit: |
SDL_math2 |
| Class: |
none |
| Declaration: |
TKnnWMode = (kwmGauss, kwmAverage, kwmMedian, kwmLinear, kwmBiQuad); |
The type declaration TKnnWMode defines the available weighting modes used in the procedure EstimateByKNN.
The weighting modes cossrespond to the following interpolation schemes:
| Weighting Mode |
Interpolation Scheme |
| kwmGauss |
The estimated values are calculated by the weighted sum of the k nearest neighbors. The weights depend non-linearly (Gaussian kernel) on the distance between the value to be estimated and the k nearest neighbors. |
| kwmAverage |
The estimated value is equal to the average of the values of the k nearest neighbors. |
| kwmMedian |
The estimated value equals the median of the values of the k nearest neighbors. |
| kwmLinear |
The estimated values are calculated by the weighted sum of the k nearest neighbors. The weights depend linearly on the distance between the value to be estimated and the k nearest neighbors. |
| kwmBiQuad |
The estimated value is obtained by a multiple linear regression which is based on the k nearest neighbors, the quadratic terms and the cross-terms of all involved variables. |
|
MathPack 
