About This Course

Get a basic understanding of the course, its prerequisites, and its target audience.

What time series are and why they matter

Consider the following industry examples:

  • A logistics company needs to optimize its transportation operations during the holiday season to meet the increased demand for package deliveries.

  • A telecommunications provider wants to forecast call volumes during major sporting events to ensure sufficient staffing and infrastructure for customer support.

  • An energy company must predict electricity demand during extreme weather conditions, such as heatwaves or cold spells, to optimize power generation and distribution.

It is clear from these three examples that seasonality, trends, and sporadic events have a direct impact on business operations, and companies need to anticipate and adapt to them.

Sequences of time-indexed observations are called time series. They can be found in many domains: finance, climate science, engineering, and, more broadly, any field that involves collecting data at different points in time. This sequential nature is not only the key to explaining the data, but it also adds predictability. And predictability is what we are after. With a solid understanding of past behavior and future outcomes, companies can make informed decisions and gain a competitive edge in many industries.

Electricity demand in the UK—an example of time series
Electricity demand in the UK—an example of time series

In this course, we’ll learn how to analyze univariate time series. We’ll see how to exploit the temporal dimension of this type of data to uncover business insights and make predictions.

Our approach

Theory and practice are equally important in this course. Each lesson introduces a concept in univariate time series and explains the rationale behind it. The explanations are meant to be very intuitive, with abundant graphs and, occasionally, simple mathematical formulae.

To put these concepts into practice, we include Python code snippets in all lessons. Most of them utilize a real-world dataset that will be the leitmotif of the course: A series of average temperatures in San Francisco (see chart below). On top of that, we'll familiarize ourselves with the statsmodels library, which provides powerful tools for time series analysis in Python. This will ensure that we get hands-on experience with every single concept.

Average temperatures in San Francisco
Average temperatures in San Francisco

By the end of this course, you’ll have a solid foundation in univariate time series analysis. You will be equipped with the necessary skills to explore, model, and forecast time series data using Python.

Course prerequisites

To follow this course, it is recommended that you have a basic understanding of statistics. In particular, you should be comfortable with concepts such as:

  • Statistical distributions, like the normal distribution.

  • Statistical indices, like the correlation index.

  • Statistical tools, such as hypothesis tests.

In terms of coding, you should at least be familiar with the basics of Python for data science. Libraries such as pandas, NumPy, and Matplotlib will be used extensively.

Intended audience

This course is designed for data scientists, analysts, researchers, and professionals from various domains who want to develop a strong foundation in time series analysis using Python. Whether you work in finance, climate science, engineering, economics, social sciences, or any field that involves working with time-dependent data, this course will equip you with the necessary skills to apply rigorous time series analysis to business problems.