Transformation of a Free-Wilson Matrix into Fourier Coefficients

Warning

This publication doesn't include Institute of Computer Science. It includes Faculty of Science. Official publication website can be found on muni.cz.

Authors	HOLÍK Miroslav HALÁMEK Josef
Year of publication	2001
Type	Article in Periodical
Magazine / Source	Quant. Struct.-Act. Relat.
MU Faculty or unit	Faculty of Science
Citation
Field	Physical chemistry and theoretical chemistry
Keywords	Free-Wilson matrix; Fourier transform; multivariate regression; standard error of prediction; statistical importance of the regression
Description	Fourier Transform is suggested as a way to change site and substituent oriented variables into new only latently dependent ones. It is possible to find out a model with a low standard error of prediction, SEP%, of activity for not yet prepared compounds. Testing power of the correlation coefficient is improved by calculating a probability, PR, that the regression is made only by chance.
Related projects:	Structure and character of bonding, properties and analysis of synthetic and natural molecular ensembles