What if I told you that you can build an ARIMA or SARIMA model automatically? 🤯
👇👇👇
#TimeSeries #Python #DataScience #MachineLearning
So far we've seen how to estimate the parameters and how to get the optimal model. ✨
But this could be done faster with auto ARIMA! 🤦♂️
I'll show you how 👇
First, import all libraries 📚
Get stock data and format it appropriately.
Build your auto ARIMA model! ⚒️
The optimal model seems to be (0, 1, 1)
This model should not be too complex and have the best forecast.
Check the summary to verify that the coefficients are significantly different from zero.
Forecast tomorrow's price! 🔮
Sell before it's too late!! 📉 😱
As you can see auto ARIMA is very easy to use!
The optimal order is different to what we got manually.
This process is a bit subjective, especially when differences are not large and there is not a clearly better model.
Forecasting share prices is not easy, otherwise, everyone would become a millionaire easily! 😅
Follow me @daansan_ml to discover more sophisticated models, which will involve other external variables (multivariate models).
Soon I'll release the code! 👨💻
Thanks 😉
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.