Analysis of Longitudinal Forest Data on Individual-Tree and Whole-Stand Attributes Using a Stochastic Differential Equation Model

This paper focuses on individual-tree and whole-stand growth models for uneven-aged mixed-species stands in Lithuania. All the were derived using a single trivariate diffusion process defined by mixed-effect parameters stochastic differential equation describing tree diameter, potentially available area, height. The of newly developed transition probability density function estimated an approximate maximum likelihood procedure. Using relationship between multivariate univariate marginal (conditional) densities, equations to predict or forecast variables, such as height, basal stand density. results are illustrated observed dataset from 53 permanent experimental plots remeasured 1 7 times. computed statistical measures showed high predictive accuracy compared with validation data that not used find parameter estimates. implemented Maple computer algebra system.