Read/Search this Article
Abstract
When age-length data are estimated from growth rings on scale or otolith of fish, a ring-skipping error may be caused by skipping some illegible rings. This error will cause bias in the estimated parameters of growth function. In this study, we developed a maximum likelihood estimation method to estimate the growth parameters and occurrence probability of skipping-error from age-length data, which is obtained from measuring the same individual twice independently. We also examined the bias and accuracy of the method using a simulation technique. When skipping error was considered in the estimation, the bias and the root mean square error of the estimate became smaller than the estimate when skipping error was not considered. This method was robust against the measurement error for the growth function including age-length data.
Journal
- NIPPON SUISAN GAKKAISHI
-
NIPPON SUISAN GAKKAISHI 62(1), 28-31, 1996-01-15
The Japanese Society of Fisheries Science