Examples and Tutorials

Practical examples and tutorials to help you understand and apply skforecast. You can also find them in the free track Forecasting with Python.

English Español 中文

Search examples

Level

Beginner Intermediate Advanced

Topic

Single series Global models Statistical models Deep learning Foundation models Probabilistic Interpretability Feature engineering Anomaly detection Performance Visualization Case study Tools

Showing all 38 examples

Getting Started: Fundamental Forecasting

This section provides essential tutorials for users who are just getting started with time series forecasting. These examples cover the most fundamental models and techniques to help you build a strong foundation in forecasting.

• Skforecast: time series forecasting with machine learning

  Build your first machine learning forecaster end to end, from lags to backtesting.

  BeginnerSingle series

• Agentic forecasting with skforecast-AI

  An AI assistant that profiles your series, picks and validates a model, and returns the runnable skforecast script behind the forecast.

  IntermediateTools

• Skforecast Studio: time series forecasting made simple

  Build forecasting models visually, no code required, and export production-ready skforecast Python code.

  BeginnerTools

• Forecasting with statistical models

  Overview of the classical statistical models available through ForecasterStats.

  BeginnerSingle seriesStatistical models

• ARIMA and SARIMAX models

  Fit, tune and validate ARIMA and SARIMAX models, including automatic order selection.

  BeginnerStatistical models

• Forecasting with gradient boosting: XGBoost, LightGBM and CatBoost

  Compare the three main gradient boosting libraries on the same forecasting problem.

  BeginnerSingle series

• Forecasting of categorical time series

  Predict the next category instead of the next value, using a classifier as the estimator.

  IntermediateSingle series

• Visualizing time series data

  Plots that reveal trend, seasonality and autocorrelation before you model anything.

  BeginnerVisualization

• Exponential smoothing models

  ETS and Holt-Winters models for series with trend and seasonality.

  BeginnerStatistical models

• ARAR forecasting models in Python

  An automated alternative to ARIMA that shortens long memory before modelling the remaining signal.

  IntermediateStatistical models

Global Models: Multi-Series Forecasting

These tutorials focus on global models and multi-series forecasting, where you can explore the use of techniques that handle multiple time series simultaneously and compare performance across different forecasting approaches.

• Global Forecasting Models I: Multi-series forecasting

  Train a single model on many series at once and predict them all.

  IntermediateGlobal models

• Global Forecasting Models II: Comparative analysis of single and multi-series forecasting

  When does one global model beat one model per series? A side-by-side comparison.

  IntermediateGlobal models

• Global Forecasting Models III: Modeling thousand time series with a single global model

  Scale to thousands of series while keeping training time and memory under control.

  AdvancedGlobal modelsPerformance

• Global Forecasting Models IV: A step by step guide using Kaggle sticker sales data

  A complete walkthrough of a Kaggle forecasting competition with a global model.

  IntermediateGlobal modelsCase study

• Global Forecasting Models V: The M5 accuracy competition

  What the winning M5 solutions teach us about global forecasting models.

  AdvancedGlobal modelsCase study

• Forecasting with deep learning

  RNN and LSTM forecasters with ForecasterRnn, from architecture to predictions.

  IntermediateGlobal modelsDeep learning

• Forecasting at scale with deep learning

  Train a single neural network over many series without losing per-series accuracy.

  AdvancedGlobal modelsDeep learningPerformance

• Clustering time series to improve forecasting

  Group similar series first, then fit one global model per cluster.

  AdvancedGlobal modelsFeature engineering

Advanced Techniques: Beyond Basic Models

For experienced users looking to deepen their forecasting skills, this section provides advanced techniques, including probabilistic forecasting, handling missing values, and more sophisticated ensemble methods.

• Forecasting with foundation models

  Zero-shot forecasting with pre-trained models such as Chronos, TimesFM and Moirai.

  IntermediateFoundation models

• Probabilistic forecasting with machine learning

  Prediction intervals and quantiles via bootstrapping, conformal methods and quantile regression.

  AdvancedProbabilistic

• Probabilistic forecasting: prediction intervals for multi-step forecasting

  Prediction intervals for multi-step horizons, balancing coverage against interval width.

  AdvancedProbabilistic

• Continuous Ranked Probability Score (CRPS) in probabilistic forecasting

  Evaluate probabilistic forecasts with CRPS, the metric that scores the whole predicted distribution.

  AdvancedProbabilistic

• Interpretable forecasting models

  Explain what drives a forecast using feature importance and SHAP values.

  IntermediateInterpretability

• Data drift detection in time series forecasting

  Spot when production data has moved away from what the model was trained on.

  AdvancedAnomaly detection

• Modelling time series trend with tree based models

  Why trees cannot extrapolate a trend, and the differentiation and detrending that fix it.

  IntermediateSingle seriesFeature engineering

• Forecasting time series with missing values

  Strategies for gaps and NaNs, from imputation to NaN-tolerant estimators.

  IntermediateFeature engineering

• Stacking ensemble of machine learning models to improve forecasting

  Combine several forecasters into a meta-model that beats each one alone.

  AdvancedSingle series

• Time series anomaly detection

  Use forecast errors and prediction intervals to flag anomalous observations.

  AdvancedAnomaly detection

• Data leakage in pre-trained forecasting models

  How pre-training data can contaminate a zero-shot benchmark, and how to check for it.

  AdvancedFoundation models

• Accelerate forecasting models with GPU

  Move gradient boosting training onto a GPU and measure what you actually gain.

  AdvancedPerformance

• Accelerate forecasting models with Intel scikit-learn-intelex

  Speed up scikit-learn estimators on Intel CPUs without changing your code.

  AdvancedPerformance

• Forecasting with XGBoost

  A focused guide to using XGBoost as the estimator inside a forecaster.

  IntermediateSingle series

• Forecasting with LightGBM

  A focused guide to using LightGBM as the estimator inside a forecaster.

  IntermediateSingle series

Real-World Challenges and Case Studies

This section includes real-world applications of time series forecasting to tackle specific challenges, such as forecasting energy demand, web traffic, and even cryptocurrency prices. Learn how to apply forecasting techniques to practical use cases.

• Forecasting energy demand with machine learning

  Hourly electricity demand with calendar and weather exogenous variables.

  IntermediateCase study

• Forecasting web traffic with machine learning and Python

  Daily visits to a website, including weekly seasonality and special events.

  IntermediateCase study

• Intermittent demand forecasting

  Forecasting sparse series that are zero most of the time.

  AdvancedCase study

• Mitigating the impact of covid on forecasting models

  Down-weight or exclude anomalous periods with weighted time series forecasting.

  IntermediateFeature engineeringCase study

• Bitcoin price prediction with Python

  A cryptocurrency case study, and an honest look at how well it can work.

  BeginnerCase study