SEMI-BLIND ROBUST IDENTIFICATION/MODEL (IN)VALIDATION WITH APPLICATIONS TO MACRO-ECONOMIC MODELLING

https://doi.org/10.3182/20050703-6-CZ-1902.00149Get rights and content

Abstract

This paper addresses the problems of worst-case identification and model invalidation of systems subject to unknown initial conditions. While in principle these problems lead to non-convex Bilinear Matrix Inequalities (BMIs), we show that tractable convex relaxations are readily available. The potential of these techniques is illustrated by identifying and validating a subsystem of the macro-economy relating inflation and interest-rates.

Keywords

Robust Identification
Model (In)Validation
Macro-Economic Modelling

Cited by (0)

View Abstract