Palgrave Macmillan, 2023. — 441 p. — (Palgrave Advances in the Economics of Innovation and Technology). — ISBN: 978-3-031-35878-4.
This book is a comprehensive guide that explores the intersection of Artificial Intelligence (AI) and forecasting, providing the latest insights and trends in this rapidly evolving field.
The evolution of forecasting methodology exhibits similar surges interspersed with periods of calm or occasional retreats. However, unlike the ocean, the field of forecasting can continue its progress as earlier causal and extrapolative procedures are enhanced by machine learning techniques, which is the focus of this volume.
There has been some recent concern that Artificial Intelligence (AI) is a rogue science. Although most of the book focuses on the subset of AI that is Machine Learning (ML), the authors embrace AI in its broadest context, to the extent of a preface written by ChatGPT. (Hey, Chat, for your information you can’t rhyme harness with a harness, try farness.).
After a broad overview of AI related to forecasting, most chapters provide state-of-the-art discussions of the impact of ML on forecasting activity relating to time series analysis, global forecasting models, large data analysis, combining forecasts, and model selection. Even newer topics such as concept drift and meta-learning sent me a googling for definitions. The remaining chapters include case studies in economics and operations research; the finale is a chapter on Forecast Value Added (FVA), a down-to-earth method for determining whether your attempts to add value by modifying your forecasting process are making things better (or worse!).
This book explores the intersection of Artificial Intelligence (AI) and forecasting, providing an overview of the current capabilities and potential implications of the former for the theory and practice of forecasting. It contains 14 chapters that touch on various topics, such as the concept of AI, its impact on economic decision-making, traditional and machine learning-based forecasting methods, challenges in demand forecasting, global forecasting models, including key illustrations, state-of-the-art implementations, best practices, and notable advances, meta-learning and feature-based forecasting, ensembling, deep learning, scalability in industrial and optimization applications, and forecasting performance evaluation.
Part I Artificial Intelligence: Present and FutureHuman Intelligence (HI) Versus Artificial Intelligence (AI) and Intelligence Augmentation (IA).
Expecting the Future: How AI’s Potential Performance Will Shape Current Behavior.
Part II The Status of Machine Learning Methods for Time Series and New Product ForecastingTime Series Forecasting with Statistical, Machine Learning, and Deep Learning Methods: Past, Present, and Future.
Machine Learning for New Product Forecasting.
Part III Global Forecasting ModelsForecasting with Big Data Using Global Forecasting Models.
How to Leverage Data for Time Series Forecasting with Artificial Intelligence Models: Illustrations and Guidelines for Cross-Learning.
Handling Concept Drift in Global Time Series Forecasting.
Neural Network Ensembles for Univariate Time Series Forecasting.
Part IV Meta-Learning and Feature-Based ForecastingLarge-scale time Series Forecasting with Meta-Learning.
Forecasting Large Collections of Time Series: Feature-Based Methods.
Part V Special ApplicationsDeep Learning Based Forecasting: A Case Study from the Online Fashion Industry.
The Intersection of Machine Learning with Forecasting and Optimization: Theory and Applications.
Enhanced Forecasting with LSTVAR-ANN Hybrid Model: Application in Monetary Policy and Inflation Forecasting.
The FVA Framework for Evaluating Forecasting Performance.