On this post, we’ll explore:
- Several types of time series outliers
- Prediction-based and estimation-based methods for detecting outliers
- The way to take care of unwanted outliers using substitute
Outliers are observations that deviate significantly from normal behavior.
Time series can exhibit outliers because of some unusual and non-repetitive event. These affect time series evaluation and mislead practitioners into erroneous conclusions or defective forecasts. So, identifying and coping with outliers is a key step to make sure a reliable time series modelling.
In time series, outliers are often split into two types: additive outliers and innovational outliers.
Additive Outliers
An additive outlier is an statement that exhibits an unusually high (or low) value relative to historical data.
An example of an additive outlier is the surge within the sales of a product because of a promotion or related viral content. Sometimes these outliers occur because of erroneous data collection. The additivity has to do…
