Oct 2, 2011

Workshop on Probability and Statistics in Finance

The Workshop on Probability and Statistics in Finance, sponsored by the IMS special interest group on “Finance: Probability and Statistics” (FPS) and the Statistics Department of Columbia University, took place on June 23, 2011 at Columbia University. After a continental breakfast and welcoming remarks, four speakers gave talks, punctuated by coffee breaks and lunch. A panel discussion followed their well attended talks. The workshop was organized by Xin Guo of U.C. Berkeley, Tze Leung Lai of Stanford University, and Philip Protter of Columbia University.

For more information on the recently formed special interest group FPS, including the forthcoming workshop at Berkeley/Stanford in June 2012 and the free membership sign-up, visit its URL http://lists.imstat.org/mailman/listinfo/fgfps

The Four Invited Talks
Robert Jarrow of the Cornell Business School in Ithaca, NY, kicked off the talks by returning to the classic topic of market efficiency, defined long ago by Fama in 1970. In joint work with Martin Larsson he gave it a modern interpretation involving nested families of filtrations. In this way, he provided a connection between the classic idea of market efficiency dependent on equilibrium theory and asset pricing models, with that of the more modern concept of the absence of arbitrage. In doing this, he was able to circumvent the classical conundrum of the “curse of the joint hypothesis.”

The second speaker was Arturo Kohatsu-Higa of Kyoto, Japan. His talk concerns the use of Lévy processes in the modeling of risky asset price processes; in particular, one often replaces Brownian motion as the driving noise in a stochastic differential equation with a Lévy process, and in particular a Lévy process that has “infinite activity,” which is to say enough small jumps to present serious challenges to an effective analysis. Financial quantities of interest often involve the expectation of functions, or functionals, of terminal values or paths of risky asset prices. When formulas are (almost always) not available for these quantities, one typically resorts to Monte Carlo methods, and this involves simulation issues. How one performs these simulations in general is a difficult open problem, and Professor Kohatsu-Higa presented a summary of recent work in the area, focusing on his own significant contributions, in particular those joint with S. Ortiz and P. Tankov.

After lunch the third speaker was Yingying Fan of USC. She spoke on a topic of much recent interest, involving the statistical analysis of path properties of risky asset prices observed in the market. Building on the celebrated test of Ait-Sahalia and Jacod for the determination of whether of not a risky asset price process is continuous or has jumps, she proposed a new test that has the advantage of reducing the variance, and in addition makes a contribution to the location of a jump.

The fourth and final talk before the panel discussion was by Nizar Touzi of the Ecole Polytechnique, of France. Motivated by a desire to approximate no-arbitrage bounds on the prices of exotic options, given the implied volatility curve of a given maturity, Professor Touzi presented an extension of the Monge-Kantorovitch optimal transportation problem. In his framework the mass is transported along a continuous semimartingale, and the cost of transportation depends on the drift and diffusion coefficients of the continuous semimartingale. The optimal transportation problem then minimizes the cost among all continuous semimartingales with given initial and terminal distributions.

The Panel Discussion
After a break following the last talk, a one hour panel discussion took place, with panelists Peter Carr of Morgan Stanley, Ronnie Sircar of Princeton, Steve Kou of Columbia, and Bala Rajaratnan of Stanford. Dr. Carr spoke of the possibilities of more collaboration between industry and academia and gave some concrete ideas; Professor Sircar spoke primarily about the role the new IMS group might play, and gave an illuminating presentation of the current state of the analogous group in SIAM; Professor Kou spoke of the recent mortgage scandals and how spatial statistics have a role to play in their study; and Professor Rajaratnam spoke of statistical issues related to the recent hot topic of high frequency trading data.

Want to start your own IMS Group? See http://imstat.org/groups/.
There are currently 24 groups, which include 14 journal-specific groups: AAP (Annals of Applied Probability); AIHP (Annales de l’Institut Henri Poincaré); AOAS (Annals of Applied Statistics); AOP (Annals of Probability); AOS (Annals of Statistics); BERN (Bernoulli Journal.); BNPML (Bayesian Nonparametrics, Random Partitions and Machine Learning); CBMS (NSF-CBMS Regional Conference Series in Probability and Statistics); ECP (Electronic Communications in Probability); EJP (Electronic Journal of Probability); EJS (Electronic Journal of Statistics); FAQ (to help new IMS Groups coordinators to startup a mailing list); FPS (Finance: Probability and Statistics); IMSGROUPS (announcements by IMS to IMS Group coordinators); LNMS (Lecture Notes–Monograph Series); PAS (Probability Abstract Service); PCML (Probability Community Mailing List); PS (Probability Surveys); SBG (Stochastics and Biology Group); SCPG (Southern California Probability Group); SPAS (Statistical Pan African Society Mailing list); SS (Statistics Surveys); SSP (Seminar on Stochastic Processes); STS (Statistical Science)

Share

2 Comments

  • […] It has organized its first conference: a one-day meeting held at Columbia University in June (see the report here or in the print Bulletin (October/November 2011, p5). The group is planning another meeting for […]

  • If we can use the standard deiiotvan of the sample, why can’t we use the Central Limt Theorem and traditional hypothesis testing? Why must we simulate sampling distributions? Why is one procedure or technology better than the other? The Guidelines for Assessmant and Instruction in Statistics Education (GAISE) Report cites several times that simulation will be used, but in an example they discuss (page 77) they talk about doing a certain simulation 200 times before being able to make a decision. This just is not possible to do in a timely fashion if more than one example is ever going to be done.

Leave a comment

*

Share

Welcome!

Welcome to the IMS Bulletin website! We are developing the way we communicate news and information more effectively with members. The print Bulletin is still with us (free with IMS membership), and still available as a PDF to download, but in addition, we are placing some of the news, columns and articles on this blog site, which will allow you the opportunity to interact more. We are always keen to hear from IMS members, and encourage you to write articles and reports that other IMS members would find interesting. Contact the IMS Bulletin at bulletin@imstat.org

What is “Open Forum”?

In the Open Forum, any IMS member can propose a topic for discussion. Email your subject and an opening paragraph (to bulletin@imstat.org) and we'll post it to start off the discussion. Other readers can join in the debate by commenting on the post. Search other Open Forum posts by using the Open Forum category link below. Start a discussion today!

About IMS

The Institute of Mathematical Statistics is an international scholarly society devoted to the development and dissemination of the theory and applications of statistics and probability. We have about 4,500 members around the world. Visit IMS at http://imstat.org
Latest Issue