Summary of H-method and Graphics
The H-method can be applied to any mathematical modeling of data, where data
are uncertain. It is typically advantageous to use the H-method if there are
more than 5 to 10 variables in the model. If there are more than 10 variables in
the model, the advantage of using the H-method can be considerable. In these
cases the prediction variance is typically reduced by more than 50%.
Besides providing with better predictions than traditional methods, these
methods provide with graphic procedures for analyzing different features in
data. These graphic methods extend the well-known methods and results of
Principal Component Analysis to any linear model. A paper has been written that
summarizes the H-method and shows the application of the graphic procedures to
Principal Component Analysis, Linear Regression and Ridge Regression.