基本信息 姓名：许威
部门：计算数学
职称：副教授
E-mail：wdxu@tongji.edu.cn





研究方向： 主要从事复杂金融衍生品及其投资组合定价与风险管理高性能算法的研究，其主要研究成果发表于《Journal of Economic Dynamics and Control》、《Quantitative Finance》、《Journal of Computational Finance》与《Journal of Derivatives》等国际权威期刊之上。基于多年来对金融风险管理方法与实践的研究，许威博士于2011年合著完成了一本基于巴塞尔协议II风险管理最佳实践的著作《金融风险测量和全面风险管理——理论、应用和监管》。该著作对巴塞尔协议II下银行风险管理的三大支柱的概念，理论与实践方法进行了详细的阐述，对金融从业人员理解和实现巴塞尔协议具有很大的帮助作用。在加拿大访问期间，他参加了加拿大中宏保险(Manulife)公司关于大规模变年金投资组合风险计算的项目，提出了利用机器学习技术对组合风险进行度量，极大的提高计算的效率。在业界，许威博士与光大证券、海通期货，兴证期货等金融机构有着广泛的合作，并参与了新华社主持的金融云计算平台构建的“十一五”科技部国家科技支撑计划项目，主要负责金融模型库的设计，构建与管理的工作，获得软件著作权一项，申请发明专利一项，主持制订了多项新华社金融数据与金融模型接口的技术标准。

工作经历： 2006.10-2007.05 加拿大滑铁卢大学(Waterloo) 博士后
2007.06-2010.02 加拿大滑铁卢大学 研究员
2010.3------至今 同济大学数学系 副教授

论文与出版物：
Books

陈公越，于盟，许威，《金融风险测量和全面风险管理——理论、应用和监管》，上海科学技术出版社，上海，2011.
Thomas Coleman and Wei Xu, 《Automatic Differentiation in MATLAB using ADMAT (with Applications)》，SIAM, Philadelphia, 2016.

Journal Papers
Financial Engineering

Wei Xu and Yufang Yin, Pricing American Options by Willow Tree Method under Jump-Diffusion Process, Journal of Derivatives, Vol. 22, No. 1, 46–56, 2014.
Wei Xu, Xi Chen and Thomas Coleman, The Efficient Application of Automatic Differentiation for Computing Gradients in Financial Applications, Journal of Computational Finance, Vol. 19, 71-96, 2016.
Ling Lu, Wei Xu and Zhehui Qian, Efficient Convergent Lattice Method for Asian Options Pricing with Superlinear Complexity, Journal of Computational Finance, Vol. 20 (4)，1-38, 2017.
Wei Xu, Zhiwu Hong and Chenxiang Qin, A New Sampling Strategy Willow Tree Method with application to Path-dependent Option Pricing, Quantitative Finance, Vol. 13, 861-872, 2013.
Ling Lu, Wei Xu and Zhehui Qian, Efficient Willow Tree Method for European-style and
American-style Moving Average Barrier Options Pricing, Quantitative Finance, Vol. 17, 889-906, 2017.
Wei Xu, Yuehuan Chen , Conrad Coleman and Thomas F. Coleman, Efficient Machine Learning Methods for Risk Management of Large Variable Annuity Portfolios, Journal of Economic Dynamics and Control, Vol. 87, 1-20, 2018.
Ling Lu, and Wei Xu, A Simple and Efficient Two-Factor Willow Tree Method for Convertible Bond Pricing with Stochastic Interest Rate and Default Risk, Journal of Derivatives, Vol. 25, 37-54, 2017.
G. Wang and W. Xu, A Unified Willow Tree Framework for One-Factor Short Rate Models and Applications, Journal of Derivatives, Vol. 25, 33-54, 2018.
High Performance Computing

Wei Xu, Ning Zheng and Ken Hayami, Jacobian-Free Implicit Inner-Iterative Preconditioner for Nonlinear Least Squares Problems, Journal of Scientific Computing, Vol. 68, 1055–1081, 2016.
Ling Lu, Wei Xu and Sanzheng Qiao, A Fast SVD for Multilevel Block Hankel matrix with Minimal Memory Storage, Numerical Algorithms, Volume 69, 875-891, 2015.
Gang Wu, Wei Xu and Huang Leng, Inexact and Incremental Bilinear Lanczos Components Algorithms for High Dimensionality Reduction and Image Reconstruction, Pattern Recognition, 48, 244-263, 2015.
Wei Xu and Wei Li, An efficient preconditioner construction for Newton-GMRES method with application to power flow study, IEEE Transaction on Power Systems, Vol. 28, 4173 - 4180, 2013
Wei Xu and Thomas Coleman, Efficient(Partial) Determination of Derivative Matrices via Automatic Differentiation, SIAM Scientific Computing, 35, A1398-A1416, 2013.
Wei Xu and Thomas Coleman, Solving Nonlinear Equations with the Newton-Krylov Method Based on Automatic Differentiation, Optimization Methods and Software, Vol. 29, 88-101, 2014
Wei Xu, Bangyi Zheng and Jiantong Zhang, Ef?cient restarted radius bisection algorithm for sphere decoding with applications to multiple-in multiple-out systems, IET Communications, Vol. 6, 3140 – 3149, 2012.
Gang Wu, Wei Xu, Ying Zhang and Yimin Wei, “A preconditioned conjugate gradient algorithm for GeneRank with application to microarray data mining”, Data Mining and Knowledge Discovery, 26 (2013), 27-56.
Wei Xu, Sanzheng Qiao and Jiantong Zhang, “A twisted factorization method for complex symmetric tridiagonal eigenvalue decomposition”, INFORMATION, An International Interdisciplinary Journal, 15（2012）, 1393-1409.
Wei Xu，Thomas. F. Coleman and Gang Liu. A Secant Method for Nonlinear Least-Squares Minimization, Journal of Computational Optimization and Applications, 51（2012）, 159–173.
Gang Liu, Shenqi Lu, Juping Chen and Wei Xu, A note on condition numbers with generalized inverse A_T,S⁽²⁾ and constrained linear systems, Applied Mathematics and Computation, (2010), 3199~3206.
Thomas F. Coleman and Wei Xu, “Fast (Structured) Newton Computations”, SIAM Scientific Computing，Vol. 31, (2008), 1175-1191.
Wei Xu and Sanzheng Qiao, “A twisted factorization method for symmetric SVD of a complex symmetric tridiagonal matrix”, Numerical Linear Algebra with Applications，16 (2009), 801-815.
Wei Xu and Sanzheng Qiao, “A divide-and-conquer method for Takagi factorization”, SIAM Matrix Analysis and Application, 30 (2008), 142-153.
Wei Xu, Yimin Wei and Sanzheng Qiao, “Condition numbers for structured least squares”, BIT, Vol. 46, No. 1 (2006), 203-225.
Xiaoqin Jin, Yimin Wei and Wei Xu. “A stability property of T.Chan's preconditioner”, SIAM Matrix Analysis and Application, 25(2003), 627-629.
Wei Xu and Sanzheng Qiao. A Fast SVD Algorithm for Square Hankel Matrices. Linear Algebra and Its Applications, 428 (2008), 550-563.
Wei Xu, Sanzheng Qiao, and Yimin Wei. “A Note on the Scaled Total Least Squares Problem”. Linear Algebra and Its Applications, 428 (2008), 469-478.
Yimin Wei, Naiming Zhang, Michael Ng and Wei Xu, “Tikhonov regularization for weighted TLS problems”, Applied Mathematics Letters, 20 (2007) 82-87.
Yimin Wei and Wei Xu, “Condition number of Bott-Duffin inverse and their condition numbers”, Applied Mathematics and Computation, 142 (2003), 79-97.

Proceeding Papers
Thomas F. Coleman, Xin Xiong, and Wei Xu, Using Directed Edge Separators to Increase Efficiency in the Determination of Jacobian Matrices via Automatic Differentiation, Recent Advances in Algorithmic Differentiation Lecture Notes in Computational Science and Engineering, Springer, 87 (2012), 209-219。
Sanzheng Qiao, Wei Xu, and Yimin Wei. An Algorithm for Solving Scaled Total Least Squares Problems. The Proceedings of the 2008 International Conference on Scientific Computing, Worldcomp'08. Las Vegas, NE, July 14-17, 2008, 211-217.
Thomas F. Coleman and Wei Xu, “Parallelism in Structured Newton Computations”, Parallel Computing: Architectures, Algorithms and Applications Proceedings of the International Conference ParCo 2007, edited by C. Bischof, M. Bücker, P. Gibbon, G. Joubert, T. Lippert, B. Mohr, F. Peters, 2007, 295-302.
Yimin Wei, Wei Xu, Sanzheng Qiao, and Huaian Diao, “Componentwise Condition Numbers for Generalized Matrix Inversion and Linear Least Squares”, Numerical Mathematics: A Journal of Chinese Universities. English Series. 14(2005), 277-286.
Sanzheng Qiao, Guohong Liu, and Wei Xu. “Block Lanczos Tridiagonalization of Complex Symmetric Matrices”, Advanced Signal Processing Algorithms, Architectures, and Implementations XV, edited by Franklin T. Luk, Proc. of SPIE Vol. 5910, 591010 (2005).
Yimin Wei, Wei Xu and Michael Ng, “The perturbation theory of truncated weighted SVD and its application in regularization”, Advances in Scientific Computing and Applications. Edited by Yayan Lu, Weiwei Sun and Tao Tang, Science Press, Beijing, New York, 2004, 368-376.

科研项目： 1.主持国家自然科学基金《基于自动微分的导数矩阵部分元素计算及其在非线性问题中的应用》（2012-2014）；
2.主持教育部博士点基金《新型网格缩减基与球型解码算法及其应用》（2012-2014）。
3. 主持国家自然科学基金面上项目《一类复杂金融产品投资组合的估值与监管资本计算的研究》（2018-2021）
4.参与科技部国家科技支撑计划《新闻报道指挥调度供稿与采编发综合技术及系统研发》子课题《基于云计算的国家金融数据分析与信息服务关键技术与应用》，（2012-2015），主要参与人。
5.参与国家自然科学基金《蒙特卡罗加速方法及其在金融衍生产品定价中的应用》（2012-2015）第一参与人
6.参与国家自然科学基金《一类信用衍生品的定价和优化控制研究》（2013-2016）第二参与人。

个人简介：
副教授，博导，中国注册会计师，主要从事金融工程与高性能计算等领域研究。许威博士毕业于复旦大学数学系计算数学专业，并获得理学学士与硕士学位；之后，就读于加拿大麦克马斯特大学计算与软件系，获计算机科学博士学位。2006年9月至2010年2月期间在加拿大滑铁卢大学数学学院做博士后与科研助理工作。2010年3月加入同济大学数学系，主要从事复杂金融衍生品及其投资组合定价与风险管理高性能算法的研究。在任职期间，他提出了一类改进的网格节点计算方法——改进的柳树法，该方法不仅能够快速有效的解决在布朗运动下各类金融衍生品的定价与风险管理问题，而且同样适用于跳扩散，随机波动率，Levy过程等金融随机模型。较之已有的计算方法，该方法实现更简便，计算更稳定，维护成本更低，具有极大的推广与应用价值。该研究成果主要发表于《Quantitative Finance》、《Journal of Computational Finance》、《Journal of Derivatives》与《Journal of Economic Dynamics and Control》等国际权威期刊之上。基于多年来对金融风险管理方法与实践的研究，许威博士于2011年合著完成了一本基于巴塞尔协议II风险管理最佳实践的著作《金融风险测量和全面风险管理——理论、应用和监管》。该著作对巴塞尔协议II下银行风险管理的三大支柱的概念，理论与实践方法进行了详细的阐述，对金融从业人员理解和实现巴塞尔协议具有很大的帮助作用。2014年7月至2016年4月在加拿大滑铁卢大学做访问****期间，许博士为高年级本科生和研究生开设了《投资组合优化》、《金融模型计算》等课程，并参与了该校与加拿大中宏保险(Manulife)公司关于大规模变年金投资组合风险计算的项目，提出了利用机器学习技术对组合风险进行度量，极大的提高计算的效率。在业界，许威博士与光大证券、海通期货，兴证期货等金融机构有着广泛的合作，并参与了新华社主持的金融云计算平台构建的“十一五”科技部国家科技支撑计划项目，主要负责金融模型库的设计，构建与管理的工作，获得软件著作权一项，申请发明专利一项，主持制订了多项新华社金融数据与金融模型接口的技术标准。目前，许威博士承担国家自然科学基金2项；教育部博士点基金1项，参与国家自然科学基金面上项目2项，以及科技部国家科技支撑计划项目1项，出版中英文著作各一部，并在金融工程与高性能计算领域SCI/SSCI权威期刊上发表论文近30篇。