报告人：

BanZHENG, PhD，Quantitative Researcher，Lyxor Asset Management，Société GénéraleGroup

郑坂（Ban ZHENG），博士，法国领先资产管理公司（法国兴业银行集团）研究员。巴黎综合理工大学(École Polytechnique)和法国国立统计与经济管理学院(ENSAE ParisTech)工程师，法国国立高等电信学院(Télécom ParisTech)应用数学专业博士。从事资产管理领域的研究工作，专注于金融大数据的处理，对冲基金的投资策略，指数跟踪产品的研究以及量化管理基金的策略开发。现兼任同济大学法国员工会会长及法国博效基金会秘书长。

报告摘要：

We provide a review on quantitative methods forlow-medium frequency trading and high frequency trading. For low-mediumfrequency trading, we review the different econometric estimators to extract atrend of a time series which is widely used in momentum strategies. Wedistinguish between linear and nonlinear models as well as univariate andmultivariate filtering. For high frequency trading, we introduce a multivariatepoint process describing the dynamics of the Bid and Ask price ofa financial asset. The point process is similar to a Hawkes process, withadditional constraints on its intensity corresponding to the naturalordering of the best Bid and Ask prices. We study this process in thespecial case where the fertility function is exponential so that the process isentirely described by an underlying Markov chain including the constraintvariable. Natural, explicit conditions on the parameters are establishedthat ensure the ergodicity of the chain. Moreover, scaling limitsare derived for the integrated point process.

此报告主要基于两篇文章:“ Trend Filtering Methods For Momentum Strategies” 和 “Modelling Bid and Ask Prices Using Constrained HawkesProcesses: Ergodicity and Scaling Limit”。

主持人： 林则夫 教授，12BET副经理

时 间：2014年4月21日 18:30-20:00

地点：12BET学术会堂702会议室

欢迎校内外师生和同行参加！