Departement Elektrotechniek ESAT - SISTA / TR 2004 - 44 Total Least Squares and Errors - in - Variables Modeling : Bridging the Gap between Statistics , Computational Mathematics and Engineering 1

The main purpose of this paper is to present an overview of the progress of a modeling technique which is known as Total Least Squares (TLS) in computational mathematics and engineering, and as Errors-InVariables (EIV) modeling or orthogonal regression in the statistical community. The basic concepts of TLS and EIV modeling are presented. In particular, it is shown how the seemingly different linear algebraic approach of TLS, as studied in computational mathematics and applied in diverse engineering fields, is related to EIV regression, as studied in the field of statistics. Computational methods, as well as the main algebraic, sensitivity and statistical properties of the estimators, are discussed. Furthermore, generalizations of the basic concept of TLS and EIV modeling, such as structured TLS, Lp approximations, nonlinear and polynomial EIV, are introduced and applications of the technique in engineering are overviewed.