Visualization of Correlation Tables by Positive/Negative Threshold for Coefficients Significance

This work introduces a novel approach to the visualization of correlation tables, guided by positive and negative significance thresholds of correlation coefficients. Traditional methods for visualizing correlation matrices often rely on heuristic color schemes, which lack a robust analytical foundation. To address this limitation, we propose a method that constructs and analyzes an ordered sequence of so called momentums, separating positive and negative correlations based on their significance. By leveraging mathematical principles of an optimal solutions, this approach enhances the clarity and interpretability of correlation patterns.