Premium
Maximum sum of binary‐coded residuals (MASBR) regression as a robust procedure for treatment of spectral data
Author(s) -
Wang JiHong,
Xie YuLong,
Yu RuQin
Publication year - 1995
Publication title -
journal of chemometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.47
H-Index - 92
eISSN - 1099-128X
pISSN - 0886-9383
DOI - 10.1002/cem.1180090505
Subject(s) - mathematics , binary number , regression , regression analysis , robust regression , statistics , linear regression , algorithm , arithmetic
In this paper, a novel robust regression method, the maximum sum of binary coded residuals (MASBR), is proposed. Instead of the sum of squared residuals used in least squares regression as the minimization criterion, MASBR regression maximizes the sum of binary coded residuals. MASBR regression is designed for cases where the conventional robust regression methods with breakdown points less than 50% fail. To circumvent the problem of being trapped in local optima, a stepwise‐varying acceptable error limit (SVAEL) algorithm is proposed. Both numerical simulation and treatment of real analytical data demonstrate the feasibility of MASBR regression in conjunction with the SVAEL algorithm.