Financial Data Analytics with Machine Learning, Optimization and Statistics

Written for readers who want to understand analytics in money and insurance, this volume combines three strands: classical financial statistics, machine learning methods, and the mathematics beneath them. The authors include an applied mathematician and statistician, actuaries with long professional experience, and working data analysts. Its focus is on current analytical techniques as they are applied in finance and insurance. It stresses the mathematical logic and statistical principles behind each method and ties them to everyday financial problems.

Each chapter works through derivations and proofs, particularly for the key results, and is supported by several realistic examples drawn from financial settings. The algorithms are implemented in Python and R, two languages widely used in applied research, academia and industry, so readers can build the models and run the programs on their own.

By the end, readers should be able to judge the quality of financial and insurance data and turn what they learn from it into timely financial decisions. They should also be able to use dimension reduction to strengthen supervised learning, and to select a suitable analytic tool for a given dataset depending on whether the goal is classification or regression-based prediction. The material also covers competencies assessed in several professional qualifications, including the Predictive Analytics exam from the Society of Actuaries and the Actuarial Statistics exam from the Institute and Faculty of Actuaries.

The book is intended for advanced undergraduate and graduate students in courses on financial engineering, statistics, quantitative finance, risk management, actuarial science, data science and mathematics for artificial intelligence. It is also meant for aspiring and working quantitative analysts in commercial and investment banking who want a reference on their shelves.