Using stochastic dominance to evaluate uncertain forestry investment

Stochastic Dominance (SD) is proposed as a quantitative technique for comparing uncertain forestry alternatives. The concept is appropriate for forestry investments because business and environmental factors change over the course of lengthy forest rotations.;SD is compatible with utility theory and accommodates the risk averse behavior typical of most managers. Applicability to forestry investments is reinforced by current SD usage in both financial portfolio management and in agricultural production. Forestry outcomes--financial and biological--are stochastic in nature and this stochastic nature is being incorporated in more and more production models.;A forestry application of SD was demonstrated through the selection of genetic families for tree improvement work. Based on 15-year volume and height data, top performers in volume were identified as early as age 5 (the first year that stem diameter was measured) including some families that expressed poor early growth but finished strongly at age 15. Between the two parameters, tree height and tree volume, volume was a more reliable predictor of performance at age 15. The analysis actually worked better at the earlier age when the data distributions were more compact. A microcomputer algorithm that generates stochastically efficient sets was developed and discussed for both first degree and second degree stochastic dominance.