The element-wise weighted total least-squares problem

A new technique is considered for parameter estimation in a linear measurement error model AX~B, A=A"[email protected]?, B=B"[email protected]?, A"0X"0=B"0 with row-wise independent and non-identically distributed measurement errors [email protected]?, [email protected]?. Here, A"0 and B"0 are the true values of the measurements A and B, and X"0 is the true value of the parameter X. The total least-squares method yields an inconsistent estimate of the parameter in this case. Modified total least-squares problem, called element-wise weighted total least-squares, is formulated so that it provides a consistent estimator, i.e., the estimate [email protected]^ converges to the true value X"0 as the number of measurements increases. The new estimator is a solution of an optimization problem with the parameter estimate [email protected]^ and the correction @DD=[@[email protected]], applied to the measured data D=[AB], as decision variables. An equivalent unconstrained problem is derived by minimizing analytically over the correction @DD, and an iterative algorithm for its solution, based on the first order optimality condition, is proposed. The algorithm is locally convergent with linear convergence rate. For large sample size the convergence rate tends to quadratic.

[1]  ScienceDirect Computational statistics & data analysis , 1983 .

[2]  Alexander Kukush,et al.  About the convergence of the computational algorithm For The Ew-tls Estimator , 2002 .

[3]  S. Huffel,et al.  Total Least Squares and Errors-in-Variables Modeling : Analysis, Algorithms and Applications , 2002 .

[4]  Sudhir Gupta,et al.  Statistical Regression With Measurement Error , 1999, Technometrics.

[5]  B. Moor Structured total least squares and L2 approximation problems , 1993 .

[6]  A. Satorra,et al.  Measurement Error Models , 1988 .

[7]  Paul P. Gallo Consistency of regression estimates when some variables are subject to error , 1982 .

[8]  P. Sprent,et al.  A Generalized Least‐Squares Approach to Linear Functional Relationships , 1966 .

[9]  Sabine Van Huffel,et al.  Recent advances in total least squares techniques and errors-in-variables modeling , 1997 .

[10]  I. Helland Statistical Regression with Measurement Error. Chi‐Lun Cheng and John W. Van Ness, Kendall's Library of Statistics 6, Arnold, London, 1999. No. of pages: xiv+262. Price: £35. ISBN: 0‐340‐61461‐7 , 2000 .

[11]  Sabine Van Huffel,et al.  Total least squares problem - computational aspects and analysis , 1991, Frontiers in applied mathematics.

[12]  Gene H. Golub,et al.  An analysis of the total least squares problem , 1980, Milestones in Matrix Computation.

[13]  Sabine Van Huffel,et al.  Consistency of elementwise-weighted total least squares estimator in a multivariate errors-in-variables model AX=B , 2004 .

[14]  J. Vandewalle,et al.  Analysis and properties of the generalized total least squares problem AX≈B when some or all columns in A are subject to error , 1989 .

[15]  George W. Fisher,et al.  Matrix analysis of metamorphic mineral assemblages and reactions , 1989 .

[16]  Maria Luisa Rastello,et al.  The Parametric Quadratic Form Method for Solving TLS Problems with Elementwise Weighting , 2002 .