ALGORITHM FOR FACTOR ANALYSIS IN INVERSIVE SPACE

Abstract

The aim of this research was to determine possibilities of latent dimension explication in inversive space which is usually completely ignored as a systematic and different mistake. Research design was made for PC by Borland Delphi developing tool. For the needs of this research and example, data for students described by 14 morphological variables have been processed. The results have shown that inversive space defined this way is not chaotic and that it possesses real existence which was possible to recognize and comprehend, and therefore credibly explain. Restrictions of models and algorithms only exist in the situation when inversion matrices of initial data are not Gramian, and this procedure is even possible in positive self-definite matrices. Practical implications of this research are more than explicit since they enable more detailed research on phenomena which were until now trapped in the space of unknown without possibilities of scientific clarification. Original way of research is indisputable, because there are simply no similar algorithms.