Can we integrate SARIMA and STL for anomaly detection? If so, what is the process like? Which model do we run first and what input goes into the other model?
I am fairly new to time series so pardon me if this sounds ignorant. Google did not provide me with the answers I needed hence my question.