| Abstrakt: |
The material of this article is based on the author's works devoted to the construction of the Pareto set in a two-criterion parameter estimation problem of the linear regression equation with the loss functions corresponding to urban and Chebyshev distances. It is known that the former is not sensitive to outliers, while the latter, in contrast, gravitates to them. In these works, it was shown that such a problem is reduced to a multicriteria linear programming problem and its solution is the Pareto set. This article proposes a method for checking whether an arbitrary parameter estimate meets the requirements of Pareto's law and, in the case that it does not meet the requirements of Pareto's law, determining the compromise estimates in the given senses. Moreover, all the problems formulated are reduced to computationally simple linear programming problems. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink