FITTING SURPLUS PRODUCTION MODELS - COMPARING METHODS AND MEASURING UNCERTAINTY

Three approaches are commonly used to fit surplus production models to observed data: effort-averaging methods; process-error estimators; an d observation-error estimators. We compare these approaches using real and simulated data sets, and conclude that they yield substantially d ifferent interpretations of productivity. Effort-averaging methods ass ume the stock is in equilibrium relative to the recent effort; this as sumption is rarely satisfied and usually leads to overestimation of po tential yield and optimum effort. Effort-averaging methods will almost always produce what appears to be ''reasonable'' estimates of maximum sustainable yield and optimum effort, and the r2 statistic used to ev aluate the goodness of fit can provide an unrealistic illusion of conf idence about the parameter estimates obtained. Process-error estimator s produce much less reliable estimates than observation-error estimato rs. The observation-error estimator provides the lowest estimates of m aximum sustainable yield and optimum effort and is the least biased an d the most precise (shown in Monte-Carlo trials). We suggest that obse rvation-error estimators be used when fitting surplus production model s, that effort-averaging methods be abandoned, and that process-error estimators should only be applied if simulation studies and practical experience suggest that they will be superior to observation-error est imators.