Pages: 177-192
Published: 30.09.2017
Abstract: The proof of the theorem on the existence of infinite sets of -samples simulated within framework of the inverse model of the principal component analysis (IM PCA) [2-5,8] with n=2 is given in the article. IM PCA was developed for any integer value of n greater than 2:n>2 [2-5,8]. Within the framework of the IM PCA, the inverse problem of the principal component analysis (IP PCA) having an infinite set of associated solutions--samples Z(t, mn, 1-dimensional z-variables in which is modeled and histogramically adequate to 1-dimensional z-variables in a real multidimen sional sample [8,12-16].It is shown how to quantify the effect of the coefficient r12=r21=r on the f-parameters f2,…,f6 [2-5,8] of the spectrum nn of the correlation matrix R()nn and on the elements of the matrices 22,C()22, the multidimensional -samples Y(t)m2,Z(t,)m2, t=1,…,кt<,=1,…,k<.
Key words: inverse model jf the principal component analysis, -sample
|
