12 Robust Statistical Methods for Dealing With Outliers

Photo by Daniel K Cheung on Unsplash

Robust statistical methods are techniques that are designed to provide reliable estimates and inferences even when the data contains outliers or is not strictly normally distributed. These methods are less sensitive to extreme values and can provide more accurate results in the presence of outliers. These methods exhibit robustness by either explicitly considering the median (which is less affected by outliers than the mean) or by using weighting schemes that down-weight the influence of extreme values. When dealing with datasets that may contain outliers, applying these robust statistical methods can lead to more reliable and stable results.

Here are some commonly used robust statistical methods:

1. Robust Measures of Central Tendency:

• Instead of using the mean (which is sensitive to outliers), use robust measures like the median.

# Assuming 'df' is your DataFrame and 'column' is the column containing numerical data
median_value = df['column'].median()

2. Robust Measures of Dispersion:

• Instead of using the standard deviation, use robust measures like the interquartile range (IQR).

# Assuming 'df' is your DataFrame and 'column' is the column…