Robust multivariate diagnostics for PLSR and application on high dimensional spectrally overlapped drug systems


Alın A., Agostinelli C., Gergov G., Katsarov P., Al-Degs Y.

JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION, vol.89, no.6, pp.966-984, 2019 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 89 Issue: 6
  • Publication Date: 2019
  • Doi Number: 10.1080/00949655.2019.1576682
  • Journal Name: JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.966-984
  • Keywords: Cook's distance, influential observations, leverage points, partial least squares regression, SIMPLS, weighted likelihood, PARTIAL LEAST-SQUARES, SIMULTANEOUS SPECTROPHOTOMETRIC DETERMINATION, OUTLIER DETECTION, PYRIDOXINE HYDROCHLORIDE, CALIBRATION, PARACETAMOL, REGRESSION, CAFFEINE, COMPLEXITY, SELECTION
  • Dokuz Eylül University Affiliated: Yes

Abstract

Statistical methods are effectively used in the evaluation of pharmaceutical formulations instead of laborious liquid chromatography. However, signal overlapping, nonlinearity, multicollinearity and presence of outliers deteriorate the performance of statistical methods. The Partial Least Squares Regression (PLSR) is a very popular method in the quantification of high dimensional spectrally overlapped drug formulations. The SIMPLS is the mostly used PLSR algorithm, but it is highly sensitive to outliers that also effect the diagnostics. In this paper, we propose new robust multivariate diagnostics to identify outliers, influential observations and points causing non-normality for a PLSR model. We study performances of the proposed diagnostics on two everyday use highly overlapping drug systems: Paracetamol-Caffeine and Doxylamine Succinate-Pyridoxine Hydrochloride.