Diff erence equations derived on the basis of the McDill-Amateis diff erential functions and from the integral form of the Bertalanff y, Richards, and Korf growth functions were used to model the dominant height growth of young European aspen (Populus tremula L.) stands on Athos Peninsula (northern Greece). Data from stem analysis were used for fi tting. Both numerical and graphical analyses were used to compare alternative models. Th e cross-validation approach was used to analyze the predictive ability of the models. Th e algebraic diff erence form of the diff erential function proposed by McDill and Amateis resulted in the best compromise between biological and statistical aspects and produced the most adequate site index curves. Th erefore, it is recommended for height growth prediction and site classifi cation of European aspen stands on Athos Peninsula. Th is equation is base-age invariant, so any number of points (A 1 , H 1 ) on a specifi c site curve can be used to make predictions for a given age A 2 and the predicted height H 2 will always be the same.

One of the most important statistical tools is linear regression analysis for many fields such as medical sciences, social sciences, econometrics and more. Regression techniques are commonly used for modelling the relationship between response variables and explanatory variables. In this study, inside bark tree diameter was used as the dependent variable and outside bark diameter and site type as independents. While generally it is assumed that inside and outside bark diameters are linearly correlated, linear regression application is weak in the presence of outliers. The purpose of this study was to develop a Multi-Layer Perceptron neural network model which considered significant variables from an a priori developed robust regression model. The application of robust regression could be considered in selecting the input variables in a neural network model.