Use of Mahalanobis distance for outlier detection

dvgodoy / handyspark

HandySpark - bringing pandas-like capabilities to Spark dataframes

MIT License

188 stars 24 forks source link

Use of Mahalanobis distance for outlier detection #28

Open FedericoTrifoglio opened 2 years ago

FedericoTrifoglio commented 2 years ago

Since the Mahalanobis distances are compared to a critical value using a Chi-Squared distribution, should this method only be used if the columns are all Normally distributed?