Premium
The MASCA Approach Applied to Break Multicollinearities in Quantitative Structure‐Activity Relationships
Author(s) -
Mager Peter P.
Publication year - 1983
Publication title -
biometrical journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.108
H-Index - 63
eISSN - 1521-4036
pISSN - 0323-3847
DOI - 10.1002/bimj.19830250610
Subject(s) - multicollinearity , regression analysis , econometrics , multivariate statistics , principal component analysis , principal component regression , regression , predictive power , statistics , mathematics , philosophy , epistemology
Since the regressors in multiple and multivariate regression analysis are intercorrelated, the coefficients of a regression of regressands on them are unreliable with respect to the predictive model power. Nonpredictive and predictive multicollinearity affect the ability to predict the biological activity (regressand) of novel drugs based on a function of physiochemical parameters (regressors), therefore. The nonpredictive multicollinearity can be broken by using the principal component regression analysis of the MASCA model. The working technique is exemplified on bispyridinium oxime antidotes, trimethoprim derivatives, and Taft's steric constant.