Statistical Modeling and Analysis in the Age of AI

This book offers practical implementation strategies for statistical learning and data science techniques, with fully peer-reviewed papers that embody insights and experiences gathered within the LISA 2020 Global Network. It offers a unifying framework that connects statistics, machine learning, and modern AI methods, with a strong emphasis on interpretation, uncertainty, and real-world impact.

As artificial intelligence increasingly shapes decision-making in science, industry, and public life, there is a growing need to move beyond black-box models toward transparent, reliable, and ethically grounded approaches. This book addresses that need by re-centering statistical reasoning at the heart of AI-driven analysis. The main topics covered include classical and generalized regression, tree-based and kernel methods, Bayesian modeling, deep learning and latent representations, time series and sequential models, model evaluation beyond accuracy, calibration and uncertainty assessment, explainable AI, fairness and bias, and reproducible analytical workflows. These methods are illustrated through applied case studies in health, social and civic data, environmental and sustainability analytics, and policy decision-support systems, highlighting both technical and societal dimensions.

It is intended for graduate students, researchers, educators, and practitioners in statistics, data science, machine learning, and applied AI who seek a principled foundation for modeling and analysis in an AI-driven world.