论文信息 - Discriminant Q2 (DQ2) for improved discrimination in PLSDA models

Discriminant Q2 (DQ2) for improved discrimination in PLSDA models

In this paper we introduce discriminant Q2 (DQ2) as an improvement for the Q2 value used in the validation of PLSDA models. DQ2 does not penalize class predictions beyond the class label value. With rigorous Monte Carlo simulations we show that when DQ2 is used, a smaller effect can be found statistically significant than when the standard Q2 is used.

[1] R. W. Lutz,et al. Metabolic profiling of glucuronides in human urine by LC-MS/MS and partial least-squares discriminant analysis for classification and prediction of gender. , 2006, Analytical chemistry.

[2] Age K. Smilde,et al. UvA-DARE ( Digital Academic Repository ) Assessment of PLSDA cross validation , 2008 .

[3] B. Skagerberg,et al. Predictive ability of regression models. Part I: Standard deviation of prediction errors (SDEP) , 1992 .

[4] A. Smilde,et al. Assessing the statistical validity of proteomics based biomarkers. , 2007, Analytica chimica acta.

[5] Lennart Eriksson,et al. Model validation by permutation tests: Applications to variable selection , 1996 .

[6] U. Edlund,et al. Visualization of GC/TOF-MS-based metabolomics data for identification of biochemically interesting compounds using OPLS class models. , 2008, Analytical chemistry.

[7] Harald Martens,et al. Reducing over-optimism in variable selection by cross-model validation , 2006 .