The LOESS Procedure
-
Overview
-
Getting Started
-
Syntax
-
DetailsMissing ValuesOutput Data SetsData ScalingDirect versus Interpolated Fittingk-d Trees and BlendingLocal WeightingIterative ReweightingSpecifying the Local PolynomialsSmoothing MatrixModel Degrees of FreedomStatistical Inference and Lookup Degrees of FreedomAutomatic Smoothing Parameter SelectionSparse and Approximate Degrees of Freedom ComputationScoring Data SetsODS Table NamesODS Graphics
-
Examples
- References
Smoothing Matrix
When no iterative reweighting is done, the “Smoothing Matrix” denoted by 
defines the linear relationship between the fitted and observed dependent variable values of a loess model. You can obtain
the predicted values of a loess fit from the observed values via
![\[ \hat{\mb {y}}=\bL \mb {y} \]](images/statug_loess0033.png){kind=small} |
where 
is the vector of observed values and 
is the corresponding vector of predicted values of the dependent variable. Note that is an n by n matrix, where n is the number of observations in the analysis. PROC LOESS does not explicitly form if the DFMETHOD=EXACT option is not explicitly or implicitly selected.