Theory of linear equations as applied to quantitative structure-activity correlations

Lj, Schaad; Ba, Hess

doi:10.1021/jm00215a001

Cited by 9 publications

(1 citation statement)

References 4 publications

(9 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[pED~ol = SK + E (4) where S is the Free-Wilson structural matrix, K is the regression coefficient matrix and E is the model uncertainty. The elements of K, the regression coefficients kij, are the least-squares optimum linear distribution of molecular activity onto the molecular substituents.…”

Section: (12)mentioning

confidence: 99%

The Free–Wilson paradigm redux: Significance of the Free–Wilson coefficients, insignificance of coefficient ‘uncertainities’ and statistical sins

Duewer

1990

Journal of Chemometrics

View full text Add to dashboard Cite

SUMMARYThe Free-Wilson paradigm is an established and powerful tool for quantitatively relating activity with chemical structure. Current implementations of the paradigm, however, are flawed both conceptually and in execution. As part of an attempt to more fully realize the promise of the paradigm, it was necessary to examine these limitations in detail.This report introduces a robust, theory-founded Free-Wilson implementation: stepwise principal components regression analysis (SPCRA). SPCRA is computationally superior to previous implementations but does not in itself correct their conceptual flaws.The development of SPCRA did, however, facilitate derivation of a simple and chemically significant interpretation of the Free-Wilson structure-activity model. A number of statistical aspects of this model commonly misused in previous applications are discussed at length. These discussions provide critical background for the development of an alternative implementation of the Free-Wilson paradigm.

show abstract

Section: (12)mentioning

confidence: 99%