On the tolerance approach to possibilistic nonlinear regression over interval data

Abstract

We study the tolerance-based approach to possibilistic nonlinear regression models with interval data. We provide a method for determination of interval regression parameters of the model for the crisp input - interval output case and for the interval input - interval output case.

We define two classes of nonlinear regression models for which efficient algorithms exist. We illustrate the theory by examples.

Keywords

• Interval regression
• Nonlinear regression
• Possibilistic regression
• Tolerance quotient