This repository contains analysis of churn in telephone service company (using IV and WOE), comparison of effect size and information value and quick tutorial how to use information value module (created for this analysis).
Hi. In case of data with repeated values, the bins will not all have same count. In such a case, it is better to calculate average of 'ones' in each bin by bin 'count', and use that to calculate spearman correlation (part of __generate_correct_bins function). The attached image might make it clearer. Thanks