- An easy to read by state of the art text containing a comprehensive review and analysis of existing corporate bankruptcy models, and their applications to real life data
- Covers a broad range of statistical learning models, ranging from relatively linear techniques (e.g. linear discriminant analysis) to state-of the art machine learning methods (e.g. random forests, deep learning).
- Explains the purpose, strength and limitations of respective models and frameworks, highlighting their major points of similarity and difference and would make this book a useful reference
-
Much of the corporate bankruptcy literature has relied on quite simplistic classification models but this book introduces a wide range of innovative corporate bankruptcy prediction models
Stewart Jones is Professor of Accounting at the University of Sydney Business School. He specializes in corporate financial reporting and has published extensively in the distress risk and corporate failure modelling field. His publications appear in many leading international journals, including the Accounting Review, the Review of Accounting Studies, Accounting Horizons, Journal of Business Finance and Accounting, the Journal of the Royal Statistical Society, Journal of Banking and Finance and many other leading journals. He has published over 150 scholarly research pieces, including 70 refereed articles, 10 books, and numerous book chapters, working papers, and short monographs. Stewart is currently Senior Editor of the prestigious international quarterly, Abacus.
Title: Distress Risk and Corporate Failure Modelling: The State of the Art (Routledge Advances in Management and Business Studies)
Author: Jones, Stewart
ISBN: 9781138652507
Binding:
Publisher: Taylor & Francis Ltd
Publication Date: 2022-09-15
Number of Pages: 230
Weight: 0.4492 kg
'This book provides a comprehensive and highly informative review of corporate distress and bankruptcy modelling literature. The book traces the early development of this literature from linear discriminant models that dominated bankruptcy research of the 1960s and 1970s to modern machine learning methods (such as gradient boosting machines and random forests) which have become more prevalent today. The book also provides a comprehensive illustration of different machine learning methods (such as gradient boosting machines and random forests) as well as several pointers in how to interpret and apply these models using a large international corporate bankruptcy dataset. A helpful book for all empirical researchers in academia as well as in business.'
Iftekhar Hasan, E. Gerald Corrigan Chair in International Business and Finance, Gabelli School of Business, Fordham University in New York, USA
'The corporate bankruptcy prediction literature has made rapid advances in recent years. This book provides a comprehensive and timely review of empirical research in the field. While the bankruptcy literature tends to be quite dense and mathematical, this book is very easy to read and follow. It provides a thorough but intuitive overview of a wide range of statistical learning methods used in corporate failure modelling, including multiple discriminant analysis, logistic regression, probit models, mixed logit and nested logit models, hazard models, neural networks, structural models of default and a variety of modern machine learning methods. The strengths and limitations of these methods are well illustrated and discussed throughout. This book will be a very useful compendium to anyone interested in distress risk and corporate failure modellin.'
Andreas Charitou, Professor of Accounting and Finance and Dean, School of Economics and Management, The University of Cyprus
'This is a very timely book that provides excellent coverage of the bankruptcy literature. Importantly, the discussion on machine learning methods is instructive, contemporary and relevant, given the increasingly widespread use of these methods in bankruptcy prediction and in finance and business more generally.'
Jonathan Batten, Professor of Finance, RMIT University
