{ "item_title" : "AI for Financial Risk Management", "item_author" : [" Sanket "], "item_description" : "This book is about how artificial intelligence can help risk functions keep pace. Machine learning finds patterns in loan books, market data and transactions that linear models miss. Generative AI reads, summarises and drafts the documents on which so much risk work depends: credit memos, loss-event narratives, regulations and policies. Agentic AI combines models and tools into systems that can monitor, investigate and recommend actions across many steps. Used well, these methods help institutions see risks earlier, measure them better and respond faster. AI also creates risk. Models can be biased, unstable or wrong in ways that are hard to see. Language models can produce fluent but unsupported answers. Agents that act on their own raise new questions about control and accountability. Supervisors have responded with the EU AI Act, updated model risk guidance in the United States and the United Kingdom, and closer scrutiny of third-party technology. A book on AI for risk management must therefore cover both how to use AI on risk and how to manage the risk of AI. This one does both. Why this bookThis volume goes deep on one function, risk management, and organises the material around the four risk types that every bank, insurer and asset manager must manage: credit, market, liquidity and operational risk. Within each type, the chapters move from machine learning, through generative AI, to agentic approaches. Liquidity risk, often neglected in the AI literature, has a part of its own. Who it is forThe book is written for four groups of readers: Risk managers and chief risk officers who need to judge where AI adds value, what it costs in new risk, and what supervisors will expect.Quantitative analysts and model developers who want practical methods, from gradient-boosted PD models to neural quantile VaR and diffusion-based scenario generators.Data scientists and engineers moving into finance, who need the domain context, data and regulatory constraints that make risk modelling different.Model validators, auditors, regulators and students who need to understand how these systems work in order to challenge them.No advanced mathematics is assumed beyond basic probability, statistics and linear algebra, which Appendix A summarises. Some familiarity with Python helps with the labs. What makes it practicalEvery chapter follows the same structure: an opening vignette based on a real event or decision, learning objectives, the classical approach and its limits, AI methods, a Python lab, a case study, a regulatory and model-risk lens, key takeaways, review questions and further reading. The 16 labs use simulated data, run offline on a laptop and are printed in full, so the book needs no online repository. Appendix B lists public datasets for readers who want to apply the methods to real data. Suggested reading paths Risk manager or CROPart I in full; the parts for your risk types; Chapters 17 and 18; Chapter 20 Quant or model developerChapter 1 briefly; Chapters 2-3; the modelling chapters (5-7, 9-12, 14-15); Chapter 18 Data scientist new to financeChapters 1-2 closely for domain context; then any risk part; Appendix C for regulation Executive or board memberChapters 1 and 4; the case studies and key takeaways in each chapter; Chapters 18-20 Validator, auditor or regulatorChapters 3, 4 and 18 first; then the regulatory lens and lab of each chapter", "item_img_path" : "https://covers1.booksamillion.com/covers/bam/9/79/816/260/9798162603252_b.jpg", "price_data" : { "retail_price" : "39.00", "online_price" : "39.00", "our_price" : "39.00", "club_price" : "39.00", "savings_pct" : "0", "savings_amt" : "0.00", "club_savings_pct" : "0", "club_savings_amt" : "0.00", "discount_pct" : "10", "store_price" : "" } }

Overview

Customers Also Bought

Details