Volume no :6, Issue no: 1, March (2017)

ROBUST ESTIMATION FOR LINEAR ERRORS-IN-VARIABLES MODELS WITH HOMOSCEDASTIC MEASUREMENT ERRORS

Author's: Cuiping Guo, Junhuan Peng and Chuantao Li
Pages: [27] - [34]
Received Date: February 10, 2017
Submitted by: Jianqiang Gao.
DOI: http://dx.doi.org/10.18642/ijamml_7100121817

Abstract

Linear errors-in-variables (EIV) models with heteroscedastic measurement errors are widely adopted in applied sciences. The linear EIV model estimators, however, can be highly biased by gross errors. This paper focuses on robust estimation in linear homoscedastic EIV models. Robust estimators of the linear homoscedastic EIV models are derived from M-estimators and Lagrange multiplier method.

Keywords

errors-in-variables, homoscedastic measurement errors, robust estimation, total least squared estimators, M-estimators.