Cathy O’Neil

Colloquium Seriesbarnsley_small

December 8, 2016: Cathy O'Neil 

Weapons of Math Destruction: A Conversation with Cathy O'Neil

Event Details: Thursday, December 8, 2016; 4:00 – 5:00pm; Dodd Center Konover Auditorium. Refreshments will be served at 3:30 pm.

About the Speaker:

Cathy O’Neil earned a Ph.D. in math from Harvard, was a postdoctoral fellow in the MIT math department, and was a professor at Barnard College, where she published a number of research papers in arithmetic algebraic geometry. She then switched over to the private sector, working as a quantitative analyst for the hedge fund D.E. Shaw in the middle of the credit crisis, and then for RiskMetrics, a risk software company that assesses risk for the holdings of hedge funds and banks. She left finance in 2011 and started working as a data scientist in the New York start-up scene, building models that predicted people’s purchases and clicks. Cathy wrote Doing Data Science in 2013 and launched the Lede Program in Data Journalism at Columbia in 2014. She is a weekly guest on the Slate Money podcast, and her latest book, Weapons of Math Destruction, is about the dark side of big data.

David X. Li (Prudential)

Colloquium Seriesbarnsley_small

December 1, 2016: David X. Li (Prudential) homepage

Some of the Latest Thoughts on Credit Portfolio Modeling

Abstract:

This talk will provide an overview about the credit portfolio modeling and some of the latest thoughts. I attempt to present a fundamental modification to the current popularly used copula function approach to the credit portfolio modeling introduced by Li (2000). The original approach simply uses a copula function to create a joint survival time distribution where individual survival time distribution is already risk neutralized, and given from a single name perspective. Based on Buhlmann's equilibrium pricing model (1980) under some assumptions on the aggregate risk or the multivariate Esscher and Wang transforms we find that the covariance between each individual risk and the market or aggregate risk should be included in the measure change. In the Gaussian copula model it is shown that we simply need to adjust the asset return by subtracting an item associated with the covariance risk. This discovery allows us to theoretically link our credit portfolio modeling with our classical equity portfolio modeling in the CAPM setting. This can help us solve some practical problems we have been encountering in the credit portfolio modeling.

Event Details: Thursday, December 1, 2016; 4:00 – 5:00pm; MONT 214

About the Speaker:

Dr. David X. Li is currently the head of enterprise risk methodology and analytics at Prudential. Previously he worked at AIG as senior managing director and head of modeling for AIG Investments. He was a managing director and chief risk officer at CICC where he was responsible for the risk management group, quantitative analytics group and new product group. Dr. Li served as Global Head of Credit Derivative Quantitative analytics from June 2004 to April 2008. Before that, he worked at Citi Group as Global Head of Credit Derivative Research from October 2001 to May 2004, in AXA Financial as Vice President of Risk Management from March 2000 to October 2001, in the Risk Metrics Group (RMG)/J. P. Morgan as partner from January 1999 to March 2000, in CIBC as executive director in the Financial Products Group from August 1996 to December 1998, in RBC as Manager in Risk Management from May 1995 to July 1996, and he taught at the University of Manitoba as assistant professor in actuarial science/finance from July 1994 to May 1995.

Dr. Li obtained his PhD degree in Statistics and master's degree in actuarial science from the University of Waterloo in 1995 and 1993 respectively, master's degree in business administration (MBA) from Laval University in 1991 and master's degree in economics from Nankai University in 1987, his bachelor's degree in mathematics from Yangzhou University in 1983.

Edward W. (Jed) Frees (University of Wisconsin – Madison)

Colloquium Seriesbarnsley_small

November 10, 2016: Edward W. (Jed) Frees (University of Wisconsin - Madison) homepage

Copulas: A Tool for Modeling Dependent Insurance Risks

Abstract:

This talk reviews the copula, a probabilistic tool widely used in insurance and other disciplines; in insurance, it is primarily used to measure association between portfolio risks. In contrast, the focus of this talk is on measuring associations among risks within a portfolio with an emphasis on copula regression modeling. I set the stage for this by describing various risk control mechanisms that the insurer has at its disposal and use this platform for describing the types of associations that are of a concern to insurers. I discuss difficulties when faced with discrete data and also introduce a new measure for advising portfolio managers on the amount of risk to retain. Use of these new tools are illustrated using data from the Local Government Property Insurance Fund, a government pool that provides building and contents as well as motor vehicle coverages to governmental entities in the state of Wisconsin.

Event Details: Thursday, November 10, 2016; 4:00 – 5:00pm; MONT 214

About the Speaker:

Edward W. (Jed) Frees is a professor in the Risk and Insurance Department of the Wisconsin School of Business. He is the Hickman Larson Chair of Actuarial Science. His research interests are in actuarial science, regression and business forecasting, and panel data. Frees is a Fellow of both the Society of Actuaries and the American Statistical Association. Frees is the only individual to be a Fellow of both organizations. Prior to earning his Ph.D., he was employed by M& R Services (a Seattle actuarial and software consulting firm), John Eriksen's & Partners (a New Zealand actuarial consulting firm), and the United Kingdom's Government Actuaries Department. In addition, in 1989-1990 he was a visiting principal researcher at the U. S. Bureau of the Census. His home has been UW-Madison since 1983, where he teaches courses in statistics and actuarial science. Frees received his Ph.D. in mathematical statistics from the University of North Carolina at Chapel Hill.

Ioannis Karatzas (Columbia) – CANCELLED

Colloquium Series - CANCELLED

Moshe Milevsky (York University)

Colloquium Seriesbarnsley_small

October 20, 2016: Moshe Milevsky (York University) homepage

Slides

The Seven Most Important Equations for Your Retirement: how to use them and the stories behind them

Abstract:

As this title suggests, in this presentation I will provide an overview of the 7 most important equations or mathematical concepts that (I believe) everyone should know before they reach their retirement years. In particular, I will discuss and review work by Leonardo Fibbonacci (interest rates), Benjamin Gompertz (mortality modeling), Edmond Halley (pension valuation), Irving Fisher (optimal spending), Paul Samuelson (investment allocation), Solomon Huebner (life insurance) and Andrei Kolmogorov (ruin probabilities). The narrative will also weave in the biography and life story of the people behind these ideas, many of whom are colourful characters in their own right.

Think of it as the 50-minute version of the course: Retirement Calculus for Poets!barnsley_small

Event Details: Thursday, October 20, 2016; 4:00 – 5:00pm; MONT 214

About the Speaker:

Moshe A. Milevsky is a tenured professor and chair of the Department of Finance at the Schulich School of Business – and member of the graduate faculty of Mathematics and Statistics -- at York University in Toronto, Canada. He has published 12 books and over sixty peer-reviewed papers in various journals, including Insurance: Mathematics and Economics, Mathematical Finance, ASTIN, Journal of Risk and Insurance, Journal of Financial and Quantitative Analysis, Journal of Derivatives, Journal of Portfolio Management as well as the Financial Analysts Journal, of which he is a member of the advisory council. In addition to academic work he has written hundreds of newspaper and magazine articles, including for the Wall Street Journal, Globe and Mail and the National Post, where he received two (Canadian) national magazine awards. His best-selling books have been translated into six languages and he has delivered over 1,000 presentations and lectures around the world. Moshe is also a fin-tech entrepreneur with a number of U.S. patents, having recently sold his QWeMA software company to CANNEX Financial Exchanges. Last year Investment Advisor magazine (in the U.S.) named him one of the 35 most influential people during the last 35 years in the financial advice business. Moshe currently lives in Toronto with his wife and four daughters, but grew-up in Latin America and the Middle East.

Dmitry Kramkov (Carnegie Mellon University)

Colloquium Seriesbarnsley_small

October 13, 2016: Dmitry Kramkov (Carnegie Mellon University) homepage

On necessary and sufficient conditions in the problem of optimal investment

Abstract:

The problem of optimal investment occupies a central role in financial theory and practice. The goal of the talk is introduce relevant mathematical techniques and formulate exact answers to basic questions of the theory such as the existence the optimal investment strategy.

Event Details: Thursday, October 13, 2016; 4:00 – 5:00pm; Andre Schenker Lecture Hall (SCHN) 151

About the Speaker:

Dmitry Kramkov is Mellon College of Science Professor of Mathematical Finance. In 1996 he received a prize of the Second European Congress of Mathematics in Budapest for his research on statistics and mathematical finance. From 1997 to 2000 Dr. Kramkov worked for Tokyo-Mitsubishi International in London, where he was the Acting Head of Research and Product Development. His main responsibility was the evaluation of complex derivative contracts. Dr. Kramkov currently serves as an Associate Editor of the journal of Finance and Stochastics. He has an affiliation with the University of Oxford, where he is a member of Man-Oxford Institute for Quantitative Finance.

Professor Kramkov's main area of research is mathematical finance. His most cited works are related to optional decomposition theorem (known also as Kramkov’s optional decomposition theorem), necessary and sufficient conditions for solvability of optimal investment problem (with Walter Schachermayer), and utility based valuation (with Julien Hugonnier, Walter Schachermayer, and Mihai Sirbu). His current research is mainly focused on equilibrium-based models of financial economics. Recent results include the development of forward and backward continuous-time price impact models with Peter Bank and Sergio Pulido and sufficient conditions for the existence of dynamic Radner equilibria (with Silviu Predoiu).

Robert C. Merton (Massachusetts Institute of Technology)

University Public Lecturebarnsley_small

October 4, 2016: Robert C. Merton (Massachusetts Institute of Technology) homepage

On the Role of Financial Innovation and Finance Science in Global Economic Growth and Development

Abstract:

A well-functioning financial system plays a critical role in economic growth and development. The sometimes-claimed dichotomy between the "real economy" and the "financial economy" is largely a fiction.

For nearly a half century, financial innovation has been a central force generally improving the global financial system with considerable economic benefit having accrued from those changes. During this same period, the field of finance underwent a remarkable transformation from a conceptual potpourri of anecdotes, rules of thumb, and manipulations of accounting data to a principles-based science, subjected to rigorous empirical examination and employing some of the most sophisticated mathematical models of uncertainty and optimization. The scientific breakthroughs in finance in this period both materially shaped, and were shaped by, the extraordinary innovations in finance practice.

Today no major financial institution in the world, including central banks, can function without the computer-based mathematical models of modern financial science and the myriad of derivative contracts and markets used to execute risk-transfer transactions as well as extract price- and risk-discovery information. In the aftermath of the financial crisis of 2008-9, however, the loss in trust in both the providers and regulators of financial services, in turn raised questions about efficacy of the changes in the financial system brought about by financial innovation and the mathematical models used to support it.

In this lecture, we explore the impact of financial innovation on the economy through a series of examples—from the past, present and forecasts for the future. These examples will illustrate how financial innovation can materially impact the real-economy, both in terms of efficiency gains and environmental sustainability. They will demonstrate how crisis can induce implementation of financial innovations, which not only address the immediate challenges created by the crisis but also provide continuing, long-term benefits to the economy. We will show how financial innovation can improve implementation of important policy and regulatory objectives by removing unnecessary and unintended, dysfunctional economic "side effects" of those policies and regulations. We will discuss issues of what is a "good model" and the implication of “inherent opacity” on the production and oversight of financial services in the future.

Event Details: Tuesday, October 4, 2016; 5:30 – 7:00pm; Laurel Hall 102

About the Speaker:

Robert C. Merton is School of Management Distinguished Professor of Finance at MIT Sloan School of Management and University Professor Emeritus at Harvard University. He was George Fisher Baker Professor of Business Administration (1988–98) and John and Natty McArthur University Professor (1998–2010) at Harvard Business School. He received a Ph.D. in Economics from MIT in 1970, then served on the finance faculty at the Sloan School until 1988 as J.C. Penney Professor of Management. He is Resident Scientist at Dimensional Holdings, Inc. Merton received the Alfred Nobel Memorial Prize in Economic Sciences in 1997 for a new method to determine the value of derivatives. He is past President of the American Finance Association, a member of the National Academy of Sciences, and a Fellow of the American Academy of Arts and Sciences. Merton’s research includes finance theory, lifecycle and retirement finance, optimal portfolio selection, capital asset pricing, pricing of derivative securities, credit risk, loan guarantees, financial innovation, the dynamics of institutional change, and improving the methods of measuring and managing macro-financial risk.

Steven E. Shreve (Carnegie Mellon University)

Meeting with Students

September 22, 2016: Steven E. Shreve (Carnegie Mellon) homepage

Quantitative Careers in the Finance Industry after the Crisis

Event Details: Thursday, September 22, 2016; 11:00 – 12:00pm; Monteith Building, Room 104

Colloquium Seriesbarnsley_small

September 22, 2016: Steven E. Shreve (Carnegie Mellon) homepage

A Diffusion Model for Limit-Order Book Evolution

Abstract:

With the movement away from the trading floor to electronic exchanges and the accompanying substantial increase in the volume of order submission has come the need for tractable mathematical models of the evolution of the limit-order book. The problem is inherently high dimensional, and any realistic description of order flows must have them depend on the state of the limit-order book.  Poisson process models for the evolution of the limit-order book have been proposed, but the analysis of these is either difficult or impossible.  In this talk, we show how diffusion scaling of a simple Poisson model, inspired by queueing theory, can lead to a rich yet tractable diffusion model for the evolution of the limit-order book.  We then show how to compute the probability of up and down price moves and the time between price changes in this model.  This is joint work with Chris Almost, John Lehoczky and Xiaofeng Yu.

Event Details: Thursday, September 22, 2016; 4:00 – 5:00pm; Andre Schenker Lecture Hall (SCHN) 151

About the Speaker:

Steven Shreve is the Orion Hoch Professor of Mathematical Sciences at Carnegie Mellon University. He has authored and co-authored five books on stochastic analysis and mathematical finance. In particular, his textbook ``Stochastic Calculus for Finance’’ is used by numerous graduate programs in quantitative finance including the University of Connecticut’s Professional Master’s in Applied Financial Mathematics. The book was voted ``Best New Book in Quantitative Finance’’ in 2004 by members of Wilmots website.

Professor Shreve's research has followed two tracks. The first has been problems in financial mathematics, including models for derivative securities, utility maximization (especially in the presence of transaction costs), optimal execution of large financial transactions, and the principal agent problem of how a bank should compensate its traders. In all these cases, continuous-time models using stochastic calculus are constructed and analyzed. A second activity has been modeling of queueing systems in heavy traffic when tasks have deadlines for completion. Although queues are intrinsically discrete-event systems, when in heavy traffic, the queue lengths can profitably be approximated by diffusions. If tasks have attributes, such as lead times until deadlines expire, the approximation is a measure-valued diffusion. In a series of papers with multiple co-authors, Professor Shreve has determined the limiting measure-valued diffusion processes obtained in a variety of queueing systems. These two threads have recently come together in the construction of diffusion approximations for limit-order books that govern trading on electronic exchanges