Multiple criteria estimation procedure for nonlinear regression models

The authors consider the problem of estimating the parameters of the nonlinear regression models. For this purpose, the least squares criterion, the minimum sum of absolute errors criterion and the minimax criterion have been used. In many practical problems the choice of the ‘best criterion’ for estimating the parameters of a nonlinear regression model is not very clear. Furthermore, it may be desirable to consider more than one criterion simultaneously rather than using one single estimation procedure. Therefore the authors propose multiple criteria formulation of the problem and illustrate one of these with a simple numerical example.