Adjusting Dynamic Models to Improve Their Predictive Ability

Dynamic models are a valuable research tool for analysis and projection. However, cumulative error may build up sufficiently to make the model virtually useless for either purpose. Several adjustments in dynamic models intended to improve their predictive ability are presented in this paper along with criteria for making these adjustments. Examples of these adjustments are shown as they were used in developing a semi‐annual model of the livestock‐meat economy. The research implications for adjustments in models after validation to simulate results of proposed policies or changes in the economic parameters of the model are examined.