Chapter 5. Working with Time Series Data

I (Thomas) remember the first time someone told me there is a hidden arrow within the FedEx logo. It blew my mind to realize that something I had known for as long as I could remember contained a deeper meaning hidden in plain sight. There are some features within Power BI that remind me of this same story. I had been using line charts in Power BI for a while before learning that with just a few more clicks, I could quickly generate more advanced analytics. This chapter is dedicated to some of the AI features seemingly hidden within common visualizations. After exploring these new abilities, it will be difficult to see these visualizations the same way again—just like the FedEx logo.

We continue the discussion from Chapter 3 about ways a business can use artificial intelligence for better decision making. Now we dive into more AI features included in the free Desktop version of Power BI. This chapter explores ways an organization can utilize the AI tools of trend lines, forecasting, and anomaly detection. Before we dive into those directly, it will be helpful to cover some fundamentals around how time series data is different from other datatypes.

More Than Just Timestamps

Time series data is a type of data that measures observations of the same variable across points of time and is dependent on previous observations.

Let’s break apart that definition into some bite-size pieces. Importantly, time series data is measuring how the same object, ...

Get Artificial Intelligence with Microsoft Power BI now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.