From the course: Python for Time Series Forecasting
Join today to access over 24,800 courses taught by industry experts.
Seasonal naive method - Python Tutorial
From the course: Python for Time Series Forecasting
Seasonal naive method
- [Instructor] For the seasonal naive method, if I execute everything, we get the purple forecast, which is basically taking the last seasonal pattern and replicating it in the future. If I change the values, I place a four instead of 12, the repeating pattern is only taken from the last four steps and being replicated over time. So that's why it's very important that if there is a seasonal pattern, to put the periods related to that seasonal pattern. If it's a year, it will be 12. But if you have a quarterly frequency, which are four observations within a year, and this is repeating, you will place a four. How do we calculate this sequence? So we start by taking the last season, which is the last 12 values, and access these values as we iterate over the steps. Now, let's apply a thought exercise. If I accessed on the last season for the 11th value, I get the last one. But then if I put the 12th value, I don't get anything because there is no data here. That's why we are applying the…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.