P. Vankeerberghen et al., ROBUST REGRESSION AND OUTLIER DETECTION FOR NONLINEAR MODELS USING GENETIC ALGORITHMS, Chemometrics and intelligent laboratory systems, 28(1), 1995, pp. 73-87
Experimental data such as calibration and pharmacokinetic data can be
contaminated with outliers. Robust regression based on the calculation
of the least median of squared residuals (LMS) is robust to the prese
nce of outliers and is used for outlier detection. The original LMS pr
ogram only handles models which are linear in the parameters. A geneti
c algorithm can be used to obtain the LMS parameters for models that a
re non-linear in the parameters. In this work the feasibility of using
genetic algorithms for LMS is demonstrated by means of curved analyti
cal calibration and pharmacokinetic data contaminated with outliers.