Exploring Auto ARIMA in Python for Multiple Time Series Forecasting

Forecasting is the process of using historical data to predict future events or trends. It is a critical tool for businesses and organizations, allowing them to plan for future changes and opportunities. In this article we will discuss about auto_arima() function and how this method can help us to applying forecasting for multiple timeseries data.

Sunset view of my hometown. Captured by Aisah.

A while ago, my boss tasked me with forecasting job opening trends for the next few months. The request was not just for the total number of job openings, but also segmented by job categories and countries in our market.

Okay that’s a lot.

Not a Medium Member yet? Use this Friend link to continue read the article.

My first impression when I knew I will do a forecasting is using ARIMA with standard procedure. If we ask ChatGPT what is the steps, it would be like this (can skip if you are already mastered the ARIMA):

1. Stationarity Check: The first step in ARIMA modeling is to check for stationarity of the time series. Stationarity means that the statistical properties of the time series such as the mean and variance remain constant over time. If the time series is not stationary, it can be made stationary by taking the first or second difference or applying a seasonal differencing. The…