SB AdvisoryApril 2, 2020 update: Keep up with the latest from Stony Brook about the coronavirus situation.  More information
Skip Navigation


The Quantitative Finance Center of AMS is dedicated to connecting its faculties and students to both the academic world and the real finance world, such as Wall Street. Weekly seminars serve this purpose.

  Abstracts can be found here

Current Seminars 

QF PhD Weekly Seminars

Date:   Friday,   Nov 17th , 2017, 11:00am-12:00pm
Math Tower, Room B-148
Name:  Ruibo Yang
Nonparametric Estimators for Statistical Functionals
Abstract:  Entropies and divergences are statistical functionals which play fundamental roles in statistics and machine learning. In many applications, it is important to estimate these functionals directly from data. I am going to introduce some nonparametric estimators for statistical functionals of nonparametric distributions based on Von Mises Expansion and the notion of influence functions.
Date: Friday, Nov 10th , 2017, 11:00am-12:00pm, Math Tower, Room B-148
Name: Deepika Kharbanda, Wang Sihan and Mihail Tsonev
Title: TBA
Date: Friday, Nov 3rd , 2017, 11:00am-12:00pm
Math Tower, Room B-148
Name: Professor James Glimm
Research opportunities in Quantitative Finance
Abstract:  I ntraday Risk Measures, Term Structure of Interest, Option pricing, and Performance Attribution.
Date: Friday, Oct 27th , 2017, 11:00am-12:00pm
Math Tower, Room B-148
Bahri Haciibrahimoglu
Analysis of the Corporate Sustainability Management and Profitability: A Cybernetics Approach
Abstract: The goal of this paper is to investigate the sustainable management of the Information Systems (IS) using the Cybernetics Approach and the Viable System Model. The research objective is to determine an efficient decision-making process for ecological management and reporting. The objective is achieved by using socially responsible investment index fund criteria while analyzing these companies’ current sustainable management practices. The nature of the research questions is to find an efficient Viable System Model for adapting marketing conditions because of the environmental changes. The social index funds are a measure to observe each company’s financial performance which was based on sustainable investment factors. The economic performance of the sustainable companies that is based on Vanguard FTSE Social Index Fund (VTSFX) is analyzed to achieve this goal. Top ten most sustainable companies are selected from VTSFX for this purpose. The charts of the day to day market fluctuations are used as an indicator to measure the market moves. The consistency of the financial performance is observed to look for stable valued organizations that are based on the eco-friendly investment practices. These companies’ current sustainability reports are analyzed to determine an efficient decision making and reporting process. The suggestive changes are explained for their Viable Sustainable System Model to increase the corporate efficiency for eco-friendly management.
Keywords: Cybernetics of Sustainability, dimensions of sustainability, sustainable performance, Dow Jones Social Index Fund, sustainability framework, supply chain sustainability, Vanguard FTSE Social Index Fund, corporate sustainability report, Viable Sustainable System Model.
Date: Friday, Oct 20th, 2017, 11:00am-12:00pm
Math Tower, Room B-148
Title:  "Double spend races" (joint work with C. Grunspan)
Article:  Please click here
Abstract:  We correct the double spend race analysis given in Nakamoto's foundational Bitcoin article and give a closed-form formula for the probability of success of a double spend attack using the Regularized Incomplete Beta Function. We give a proof of the exponential decay on the number of confirmations, often cited in the literature, and find an asymptotic formula. Larger number of confirmations are necessary compared to those given by Nakamoto. We also compute the probability conditional to the known validation time of the blocks. This provides a finer risk analysis than the classical one.
Date: Friday, Oct 6th , 2017, 11:00am-12:00pm
Math Tower, Room B-148
Name: Bahri Haciibrahimoglu
Title:  Analysis of the Financial Relationship of Network Neutrality Principles
Abstract: Net neutrality is the concept that Internet Service Providers (ISPs) and governments should treat all data flow on the Internet equally. Net neutrality will impair the ability of ISPs to recover funds that they must invest in infrastructure to provide Internet services. The 2015 Open Internet Order imposed new rules for the management of broadband Internet access services. The Order defines these services as mass-market retail services that allow access to the entire Internet medium, and it includes both wired and wireless services. This paper is intended to investigate the evolutionary share market value effect of the Internet Service Providers on the network neutrality, with the emphasis on the net neutrality principles. The objective is achieved by using quantitative analysis of statistical data and content analysis of the economics of net neutrality. The Netflix ISP Speed Index has been used to determine ISP network traffic speed since Netflix is one of the biggest content providers worldwide. One year of network traffic speed graph is analyzed to oversee major contributors within. These providers’ day-to-day stock market value is downloaded from the Yahoo! Finance and their graphical data are analyzed for a 15-month period. The content analysis of economic papers is added to determine some of the qualitative aspects of the financial perspectives of net neutrality. The research is limited to findings from Netflix ISP Speed Index Graph, Yahoo! Finance intraday stock value, and some of the perspectives that will be extracted from the scholar economic papers. Its intention is to analyze the financial relationship of net neutrality. The legal framework of the net neutrality is excluded from the research.
Date: Friday, September 29 , 2017, 11:00am-12:00pm
Math Tower, Room B-148
Name: Eric Werneburg
Title: On “Simultaneous Prediction of Independent Poisson Observables”
Abstract: Random vector prediction is tedious in large dimensions, and the use of informative priors becomes necessary.  Komaki (2004) introduces a slick shrinkage prior for Poisson random vectors with independent entries and proves the resulting predictive distribution dominates that based on Jeffreys prior under the Kullback-Leibler loss when dimension is greater than 2.  We give an exposition of Komaki’s paper and perform analysis using Matlab to empirically show Komaki’s predictive method is favorable.
Date: Friday, September 22 , 2017, 11:00am-12:00pm
Math Tower, Room B-148
Name: Xingxing Ye
Title: Systemic Risk Indicators based on Nonlinear PolyModel
Abstract: The global financial market is becoming extremely interconnected and exhibits strong nonlinear contagion especially when crisis is coming, so measuring financial systemic risk in a comprehensive perspective and nonlinear way becomes necessary. In this study, we establish a large set of risk factors as the main bones of the financial market and apply nonlinear factor analysis by PolyModel to propose two systemic risk indicators which can signal early warnings of financial crises and well trace the evolution of financial crises. Through financial network analysis, theoretical simulation, empirical data analysis and final validation, we validate that our indicators are very effective in forecasting financial crises that happened during 1998 to 2017. In theoretical network analysis, we firstly describe market as a network, then decompose its risk against each risk factor in the set with three independent parts. Secondly, aggregate these three parts respectively by weighted average to form the final risk components: systemic risk, factor-specific risk and residual risk, based on which, we then propose our systemic risk indicators. In theoretical simulation, we assume the financial market as the stochastic process with the decomposed independent components above and simulate them to demonstrate the systemic risk in the market. In addition, we use both temporal return aggregation simulation and empirical data to illustrate that nonlinearity exists obviously in the market. In empirical data analysis, we choose historical weekly data of SPX Index and nearly 200 market representative risk factors and process them with 6-month moving window from 1998-2017. Firstly, we regress SPX Index against each risk factor respectively with nonlinear PolyModel which is basically implemented by the regression with Hermite polynomials basis. Secondly, we calculate our systemic risk indicators according to the formulas in the financial network analysis part and show that they can well signal early warnings of financial crises. In the final, we design investing strategies of SPX Index and its options based on the adjusted systemic risk indicators to validate our study result.
Date: Friday, September 15 2017, 11:00-Noon
Math Tower, Room B-148
Name: Chi Kong
Title: Functions of (Block) Toeplitz Matrices: High Performance Computation
Abstract: Large scale Toeplitz Matrices and asymptotic linear algebra is closely related to Operator Theory and Functional Analysis. The last few decades witnessed large progress in computations of Toeplitz and Block Toeplitz matrices, with a wide range of applications. The talk focuses on the numerical computation of functions of large scale block Toeplitz Matrices, including Matrix Inversions, Multiplications, Exponentials and Logarithms. We propose algorithms exploiting Block Toeplitz, Block Circulant and Block epsilon-Circulant matrix structures to achieve high performance , low complexity and high numerical stability. 

Date: Friday, September 8 2017, 11:00-Noon
Math Tower, Room B-148 at the Math Tower
Student: Juehui Zhang
Title: Poly-model and StressVaR on Portfolio Analysis
Abstract: In our research, we introduce a novel approach to portfolio risk estimation based on nonlinear "poly-models": a collection of nonlinear dynamic single factor models. Using this approach, we build a new risk measure called "StressVaR" which combines the notion of Value-at-Risk and of stress scenarios. StressVaR estimates the risk (potential loss) of an asset based on the long history of relevant risk factors among a very broad set of possible risk sources. We start with single asset scenario. Then we calculate the StressVaR of a portfolio based on the single asset nonlinear models. According to the StressVaR of the portfolio and the risk contribution of each component, we then try to find an approach to get an optimal portfolio.

Login to Edit