A Method for Fitting of Schaefer Model with Autoregressive of Order One

Schaefer model is one of the most popularly used surplus production models in fisheries to estimate maximum sustainable yield and the corresponding optimum fishing effort. When we employ nonlinear estimation procedures for estimating the parameters, high parameter correlations are generally observed, which is undesirable. Moreover, catcheffort fisheries data are collected during a certain time period and hence data points are generally correlated among themselves. In the present paper, utility of expected value parameters to make parameter correlations low is highlighted. An explicit expression giving catch-effort relationship for Schaefer model with autoregressive of order one using expected value parameters is also developed. This is illustrated with an example, considering the serially correlated catch-effort data observed from Gobindsagar reservoir.