From the course: Python for Time Series Forecasting
Join today to access over 24,800 courses taught by industry experts.
Model fit and forecast - Python Tutorial
From the course: Python for Time Series Forecasting
Model fit and forecast
- [Instructor] We work with the same dataset of passengers. And at the time we fit the model, pay attention on the differences between the ARIMA and SARIMAX. It comes from a different submodule within the stats model library, and we add a seasonal order to better capture the seasonal patterns in the data. From the previous lessons, if we executed the ARIMA, we got an optimization failure to converge for having many autoregressive lags in this parameter. We observe having many P-values, which are insignificant on the autoregressive component. And if we develop the SARIMAX by placing the autoregressive into the seasonal order, which means that there is a repeating pattern every 12 periods and errors repeat every 12 months. On the other hand, the ARIMA autoregressive 12 was saying, the value today depends on the value 12 steps ago and on the error from one step ago. In other words, it's not a seasonal model, it's treating lag 12 as just another lag. The ARIMA treats the 12 in…
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.