Fitting on residuals
Another method of detecting outliers from time series is fitting a classifier to the residuals from decomposition. This is also a univariate approach with the advantage of being much faster than other multivariate methods.
The apple dataset has been loaded, along with MAD estimator and the seasonal_decompose function.
This exercise is part of the course
Anomaly Detection in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
results = seasonal_decompose(apple['Volume'], period=365)
# Extract and reshape the residuals
residuals = ____
residuals = ____