Time Series Made Simple
Understand the theory. Master the code. Build better forecasts.
Time-series analysis doesn’t have to be a black box. Time Series Made Simple equips you to work with time series data effectively. Learn the theory behind forecasting methods and put those methods into practice with Python.
Whether you’re learning the fundamentals or comparing models for a real forecasting problem, the goal is the same: understand why a method works, when to use it, and how to evaluate its results.
Explore the tutorials · About this site
Learn the concepts, not just the code
Good forecasting starts with understanding the data. Each tutorial aims to build intuition first, introduce the relevant statistical ideas, and then demonstrate how to apply them in Python.
Foundations
Understand trends, seasonality, stationarity, autocorrelation, and the structure of time-series data.
Statistical forecasting
Explore moving averages, exponential smoothing, ARIMA, SARIMA, and other statistical forecasting methods.
Python in practice
Work through examples using pandas, matplotlib, statsmodels, and scikit-learn, with code that illustrates the ideas being discussed.
Model evaluation
Learn how to compare forecasts, diagnose errors, establish meaningful baselines, and decide which models are useful for a particular problem.
A practical approach to learning
The tutorials follow a consistent progression:
- Build intuition. Understand the problem and the patterns in the data.
- Explore the theory. Learn the statistical concepts behind the method.
- Implement the model. Work through a Python example.
- Evaluate the results. Examine forecast accuracy, limitations, and practical trade-offs.
The emphasis is on developing the judgment to choose and evaluate forecasting methods, rather than simply learning how to call a modeling function.
Latest tutorials
Browse the latest articles below. New articles are posted regularly, so be sure to check back soon.