… systematic investment algorithms. Thirdly, we explore whether various statistical “tricks” such as Z-scoring predictors and winsorizing values improve upon model outcomes. Our results are impactful to a variety of stakeholders in the bankrupt…
Winsorizing: Substitute outliers with the highest value that isn’t an outlier.
Using Z-scores: Replacing extreme scores with a score 3 standard deviations from the mean. A z-score of 3.29 constitutes an outlier (see 5.3.1) so we can calculate what score would give rise to a z-score of 3.29 (or perhaps 3) by rearranging the z-score equation, which gives us X = (z s) + X. All we’re doing is calculating the mean (X) and standard deviation (s) of the data and, knowing that z is 3 (or 3.29 if you want to be exact), adding three times the standard deviation to the mean.