A Bayesian hierarchical model for modeling white shrimp (Litopenaeus vannamei) growth in a commercial shrimp farm

The paper explored the Bayesian hierarchical model as a possible way to incorporate growth variability in estimating shrimp growth function to enhance forecasting accuracy, using data from 16 growout ponds of a commercial shrimp farm in Hawaii. Based on a dataset of 571 weekly growth observations, the Bayesian hierarchical model is found to fit the data better than the simple nonlinear model that neglects growth variability, with respect to the deviance information criterion, root mean squared error and mean absolute percentage error. The Bayesian hierarchical model therefore could be a promising alternative for forecasting shrimp growth in commercial aquaculture practice.