The choice of the number of factors is performed considering
that the number should be equal to or greater than three (initial
species) and by the inspection of eigenvalues and the percentage
of variance calculated by SVD analysis of the individual data
matrices (both DUV–vis-DAD and D1H-NMR-c). Table 1 shows these
values for the first 10 factors. It can see that, for the DUV–vis-DAD
data matrix, from the fourth factor the variations of the eigenvalue
are non significant. When the magnitude of the eigenvalue for the
D1H-NMR, data matrix is inspecting, it is not easy detect an inflexion
point to consider the value representative of the noise. So, we
decided retained successively different number of factor, always
higher than four, and resolve the system to analyze the solution
considering other quality parameters.
The choice of the number of factors is performed consideringthat the number should be equal to or greater than three (initialspecies) and by the inspection of eigenvalues and the percentageof variance calculated by SVD analysis of the individual datamatrices (both DUV–vis-DAD and D1H-NMR-c). Table 1 shows thesevalues for the first 10 factors. It can see that, for the DUV–vis-DADdata matrix, from the fourth factor the variations of the eigenvalueare non significant. When the magnitude of the eigenvalue for theD1H-NMR, data matrix is inspecting, it is not easy detect an inflexionpoint to consider the value representative of the noise. So, wedecided retained successively different number of factor, alwayshigher than four, and resolve the system to analyze the solutionconsidering other quality parameters.
การแปล กรุณารอสักครู่..
