Time Series & Forecasting
Predicting what comes next from data over time: trend, seasonality, ARIMA, Prophet and deep-learning forecasters.
Beginner
Start here. No background needed beyond school maths.
Forecasting: Principles and Practice (3rd edition, R)
The standard free textbook on forecasting: trend and seasonality, exponential smoothing, ARIMA, regression with time series and how to judge a forecast, with worked examples in R.
Sign up to see the source
Forecasting: Principles and Practice (Python edition)
The same textbook rewritten with Python code, so you can follow every method with pandas and statsforecast instead of R.
Sign up to see the source
Prophet: Quick Start
Fit a forecasting model with trend, yearly and weekly seasonality and holiday effects in a few lines of Python or R, then plot and check the forecast. A gentle first forecasting tool.
Sign up to see the source
Advanced
Once the basics are comfortable.
Darts: Quickstart
One Python library for many forecasting models, from ARIMA to deep learning, with backtesting, probabilistic forecasts and anomaly detection on the same time-series objects.
Sign up to see the source
StatsForecast Documentation
Fast statistical forecasting in Python: ARIMA, ETS, Theta and other models that run across thousands of series at once, with cross-validation and benchmarks.
Sign up to see the source
statsmodels: Time Series Analysis
Reference and examples for classical time-series models in Python: ARIMA and SARIMAX, exponential smoothing, VAR, state-space models and tests for stationarity.
Sign up to see the source
TensorFlow: Time Series Forecasting
A hands-on tutorial that forecasts weather data with windowed inputs, then compares dense, convolutional and recurrent (LSTM) neural networks against simple baselines.
Sign up to see the source
Expert
Research-level depth and specialist material.