Visualization of the association between variables Relevance Networks for the Liver Toxicity data were generated from the results obtained with the sPLS-reg method.
To estimate independent relevance between CTv and clinical variables, backward Akaike Information Criterion was performed, and the results are shown in Table 2.
Multivariate data analysis (MVDA) is a form of statistics that helps to understand the relationships between variables, observations, and their relevance to each other (using principal component analysis, PCA) as well as relationships between variables and responses (using partial least squares, PLS).
Since the relevance networks are visual representations of the correlations between variables, one looks for clusters or sub-networks of subsets of variables, where the edge colors indicate the nature of the correlation (positive, negative, strong or weak).
Thus, we argument and predict correlations between variables and find evidence of a certain relevance of the issue.
F-statistic is chosen to score the relevance between the genes and the classification variable.
Posterior values are between 0 and 1, where 0 indicates no relevance, 1 indicates 100% relevance between a predictor and a target variable.
Having passed through a set of rigorous validation procedures, we were able to confirm the relevance between the questionnaire items and the latent variables previously established.
Two way interactions between variables were considered.
For each data subgroup, Spearman rank correlations were used to evaluate collinearity between variables prior to conducting meta-regression; if evidence of collinearity was identified (ρ ≥ 0.8), only one of the correlated variables was modelled based on its relevance.
