Gaussian Process Models for Quantitative Finance
暫譯: 量化金融的高斯過程模型
Ludkovski, Michael, Risk, Jimmy
- 出版商: Springer
- 出版日期: 2025-03-07
- 售價: $2,320
- 貴賓價: 9.5 折 $2,204
- 語言: 英文
- 頁數: 138
- 裝訂: Quality Paper - also called trade paper
- ISBN: 3031808738
- ISBN-13: 9783031808739
海外代購書籍(需單獨結帳)
相關主題
商品描述
This book describes the diverse applications of Gaussian Process (GP) models in mathematical finance. Spurred by the transformative influence of machine learning frameworks, the text aims to integrate GP modeling into the fabric of quantitative finance. The first half of the book provides an entry point for graduate students, established researchers and quant practitioners to get acquainted with GP methodology. A systematic and rigorous introduction to both GP fundamentals and most relevant advanced techniques is given, such as kernel choice, shape-constrained GPs, and GP gradients. The second half surveys the broad spectrum of GP applications that demonstrate their versatility and relevance in quantitative finance, including parametric option pricing, GP surrogates for optimal stopping, and GPs for yield and forward curve modeling. The book includes online supplementary materials in the form of half a dozen computational Python and R notebooks that provide the reader direct illustrations of the covered material and are available via a public GitHub repository.
商品描述(中文翻譯)
本書描述了高斯過程（Gaussian Process, GP）模型在數學金融中的多樣應用。受到機器學習框架變革性影響的驅動，本文旨在將GP建模融入量化金融的基礎中。書的前半部分為研究生、資深研究人員和量化實務者提供了一個入門點，以熟悉GP方法論。系統且嚴謹地介紹了GP的基本概念以及最相關的進階技術，例如核選擇、形狀約束的GP和GP梯度。後半部分則調查了GP應用的廣泛範疇，展示了其在量化金融中的多功能性和相關性，包括參數化選擇定價、用於最佳停止的GP代理模型，以及用於收益率和遠期曲線建模的GP。書中還包括在線補充材料，形式為六個計算的Python和R筆記本，為讀者提供所涵蓋材料的直接示例，並可通過公共GitHub存儲庫獲得。
作者簡介
Mike Ludkovski is a Professor of Statistics and Applied Probability at University of California Santa Barbara. He was Department Chair during 2018-2022 and since 2016 is a Co-Director of the Center for Financial Mathematics and Actuarial Research. He has 15+ years of experience and 80+ publications in stochastic modeling of energy markets, numerical methods for stochastic control and predictive analytics. Among his current research interests are Monte Carlo techniques for optimal stopping/stochastic control, non-zero-sum stochastic games, and applications of machine learning in longevity and non-life insurance. He serves on 5+ Editorial Boards and his research has been funded by NSF, ARPA-E and Society of Actuaries. During 2015-2016 he was Chair of the SIAM Activity Group on Financial Mathematics & Engineering. He co-edited the volume on "Commodities, Energy and Environmental Finance" (2015). Ludkovski holds a Ph.D. in Operations Research and Financial Engineering from Princeton University and has held visiting positions at London School of Economics and Paris Dauphine University.
Jimmy Risk is an Assistant Professor of Mathematics and Statistics at California Polytechnic State University Pomona. He was temporary chair during Summer 2022 and has advised nine master's thesis students since taking his position in Fall 2017, several of which involving applications of Gaussian processes in modern data science including neural networks, natural language processing, and super-resolution. His education involves a Ph.D. in Statistics and Applied Probability with an emphasis in Financial Mathematics from University of California Santa Barbara, which has driven publications involving pricing and tail risk analysis using Gaussian processes to approximate conditional expectations. Additionally, Risk has an extensive actuarial science background, including developing a Gaussian process model for mortality rates, and more recently winning an open international mortality prediction contest held by the Society of Actuaries alongside Mike Ludkovski. Risk's recent research interests involve the theory and applications of Gaussian process kernels, which lie in the Reproducing Kernel Hilbert Space framework.
作者簡介(中文翻譯)
邁克·盧德科夫斯基是加州大學聖塔巴巴拉分校的統計學與應用機率教授。他在2018年至2022年間擔任系主任，自2016年以來擔任金融數學與精算研究中心的共同主任。他擁有超過15年的經驗和80篇以上的出版物，專注於能源市場的隨機建模、隨機控制的數值方法和預測分析。他目前的研究興趣包括最佳停止/隨機控制的蒙地卡羅技術、非零和隨機遊戲，以及機器學習在壽險和非壽險中的應用。他擔任5個以上編輯委員會的成員，其研究獲得了美國國家科學基金會（NSF）、高級研究計畫局（ARPA-E）和精算學會的資助。在2015年至2016年間，他擔任SIAM金融數學與工程活動小組的主席。他共同編輯了《商品、能源與環境金融》（2015）一書。盧德科夫斯基擁有普林斯頓大學的運籌學與金融工程博士學位，並曾在倫敦政治經濟學院和巴黎多芬大學擔任訪問職位。
吉米·瑞斯克是加州州立大學波莫納分校的數學與統計學助理教授。他在2022年夏季擔任臨時系主任，自2017年秋季以來指導了九名碩士論文學生，其中幾篇涉及高斯過程在現代數據科學中的應用，包括神經網絡、自然語言處理和超解析度。他的學歷包括加州大學聖塔巴巴拉分校的統計學與應用機率博士學位，專注於金融數學，這驅動了他在使用高斯過程近似條件期望的定價和尾部風險分析方面的出版物。此外，瑞斯克擁有廣泛的精算科學背景，包括為死亡率開發高斯過程模型，並且最近與邁克·盧德科夫斯基一起贏得了由精算學會舉辦的國際死亡預測比賽。瑞斯克最近的研究興趣涉及高斯過程核的理論和應用，這些核位於重現核希爾伯特空間框架內。
