A robust regression analysis of recruitment in fisheries

Variations in environmental variables and/or errors in measuring stock and recruitment often result in large and heterogeneous variations in fitting fish stock-recruitment (SR) data to a regression model. This makes the commonly used least squares (LS) method inappropriate in estimating the SR relationship. The following procedure is proposed: to identify possible outliers in fitting data to a given stock-recruitment (SR) model using the least median of the aquared orthogonal distance that is not sensitive to atypical values and requires no assumption on distribution of errors; to apply the least squares method to the SR data with defined outliers being down weighted. Examination of four sets of published field data leads the authors to suggest fitting fish SR data to suitable models using the proposed estimation method and interpreting the results with the assistance of knowledge on the relevant environmental variables and measurement errors. -from Authors