1994 Volume 60 Issue 3 Pages 359-364
The aim of this article is to consider the decision scheme of the sampling design of the survey for the maximum likelihood estimation of the stock composition in the mixed population. We evaluate the precision of the estimate taking full account of the sampling variance in the baseline populations and the mixed populations. Through numerical examples, we studied the optimal allocation of the survey effort on the baseline and mixed populations, and selection of the number and the types of the restriction enzymes in the genetic analysis.