Absolute Value Correlation Outperforms Pearson's with Outliers, Improving Data Accuracy.
The study compared two methods for measuring correlation: Pearson’s correlation coefficient and An Absolute Value correlation coefficient. An Absolute Value correlation coefficient was found to be more accurate and robust when dealing with outliers in the data. This means it gives better results even when there are unusual data points. Overall, An Absolute Value correlation coefficient is a better estimator because it has lower bias, standard error, and root mean square error compared to Pearson’s correlation coefficient.