Ah, time series forecasting. It’s the quintessential task for many data scientists, somewhat universal across various industries. This field is so valuable because if you have a crystal ball to see some key numbers ahead of time, you can use that information to get a head start and prepare for what’s coming down the pipeline.
Consider a call center: forecasting call volume allows for optimized staffing, ensuring efficient handling of customer inquiries. In retail, predicting when an item will go out of stock enables timely reordering, preventing lost sales and maximizing revenue. And of course, the holy grail of stock market prediction: if you could do this, you would be rich.
In this article, I want to show you how to do it the easy way using the awesome library sktime, the scikit-learn of time series forecasting.
Fair question! It’s kind of like asking, “Why use a fancy food processor when I have a knife and a cutting board?”. Sure, you could chop…