PROJECTION PURSUIT REGRESSION FOR MODERATE NONLINEARITIES

We present methods specially designed to be effective with moderately non-linear regression relationships. The model fitted is of the Projection Pursuit Regression (PPR) type with a smooth, non-parametric link function connecting the mean response to a linear combination of the regressors. New algorithms, close to ordinary linear regression, are developed. Considerable numerical evidence is given to substantiate the claims that the new methods outperform standard PPR when the non-linearity is moderate and the signal to noise ratio small.