This paper proposes a new method for bracketing a set S defined by nonlinear inequalities between an inner set S− and an outer set S+. Contrary to existing approaches for which S− and S+ are unions of boxes, these two sets are defined as unions of polytopes. This characterization makes it possible to describe S and compute its volume in a more accurate way than with classical methods. The resulting approach is used to quantify the influence of a given interval datum in parameter estimation, when the feasible set for the parameters is defined as the set of all parameter vectors consistent with all interval data. In order to detect potential outliers, we characterize the influence of any given datum on this set by its safely defined as the ratio between the volumes of the feasible sets computed with and without this datum. This problem amounts to computing volumes of sets as accurately as possible.
[1]
Eric Walter,et al.
Guaranteed non-linear estimation using constraint propagation on sets
,
2001
.
[2]
Eldon Hansen,et al.
Bounding the solution of interval linear equations
,
1992
.
[3]
Ramon E. Moore.
Methods and applications of interval analysis
,
1979,
SIAM studies in applied mathematics.
[4]
Ramon E. Moore.
Parameter sets for bounded-error data
,
1992
.
[5]
Luc Jaulin,et al.
Interval constraint propagation with application to bounded-error estimation
,
2000,
Autom..
[6]
Eric Walter,et al.
Set inversion via interval analysis for nonlinear bounded-error estimation
,
1993,
Autom..
[7]
Eric Walter,et al.
Light scattering data analysis via set inversion
,
1997
.