Accurate forecasting of farm performance is essential to efficient policy development and allocation of resources to farmers most in need of assistance. To accurately predict farm performance researchers need to consider the distribution(s) of critical variables beyond the mean and variance of these variables. Given that the distributions of many variables are not normal, and are often correlated, a technique to simulate farm performance must incorporate these two features. The method proposed in this study can be used to reproduce non-normally distributed variables that are correlated, these variables are then employed to predict farm performance. The simulation results show that the distributions of critical variables do impact on the measures of farm performance, and that policy advisers need to consider the skewness and kurtosis, along with the mean and variance, of distributions when analysing farm performance data and making policy advice.

Recent Conference Papers

Rural Development Centre Home Page

This page created and maintained by Ian Reeve.

Last updated on 16/10/96.

ireeve@metz.une.edu.au

Copyright 1995 University of New England, Armidale 2351
New South Wales, Australia.