Structural Vector Error Correction Modeling of Integrated Sportfishery Data

Abstract

We demonstrate how to specify and estimate a time series model that can isolate the effects of changes in fishery policy and forecast the outcome of policy changes in the context of changing climate and economic factors. The approach is illustrated with data from the headboat fishery for red snapper in the Gulf of Mexico. The initial data analysis finds that effort and harvest are cointegrated series and that effort appears to respond somewhat to past changes in harvest. This suggested a structural vector error correction model specification. Model estimation results indicate that seasonal closures directly influence both harvest and effort, whereas bag and minimum size limits only affect harvest directly. Also, climate activity has a moderate influence on this fishery, mainly via changes in effort. Model forecasts are evaluated relative to a more naïve specification using out-of-sample data and the use of the model for policy analysis is demonstrated.