A variability measure for estimates of parameters in interval data fitting
share
The paper presents a construction of a quantitative measure of variability for parameter estimates in the data fitting problem under interval uncertainty. It shows the degree of variability and ambiguity of the estimate, and the need for its introduction is dictated by non-uniqueness of answers to the problems with interval data. A substantiation of the new variability measure is given, its application and motivations are discussed. Several examples and a series of numerical tests are considered, showing the features of the new characteristic and the specifics of its use.
READ FULL TEXT