Premium
Diagnostics statistics in QSAR
Author(s) -
Mager Peter P.
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.1180090307
Subject(s) - quantitative structure–activity relationship , leverage (statistics) , outlier , multicollinearity , mathematics , statistics , applicability domain , computer science , econometrics , machine learning , regression analysis
The formal application of a Hansch analysis to a series of 3‐quinuclidinyl benzylates (QNBs) led to a ‘statistically significant’ QSAR equation. In contrast, the application of the MASCA model has shown that the design matrix is unsuitable for each QSAR analysis: one sample member is an outlier but not a high‐leverage or influential point; another one is an influential point, a high‐leverage point and an extra‐carrier point. The regressors of the design matrix are multicollinear without predictive model power. The result of such flagged observation and this type of multicollinearity is a multiple cluster correlation. The QNB series is a good example for ‘sampling artifacts’ where no practically important but artificial QSARs can be found.