Adaptive Clinical Trial Design and Simulation

         Mark Chang
         Executive Director
         AMAG Pharmaceuticals
         E-mail: mchang at amagpharma.com

         1:00-5:00pm, April 20th, 2012 at Crosstown Center Room 460/462

Summary

In this short course, we will review the basic concepts and methods for adaptive clinical trial designs, including the group sequential, sample-size re-estimation, dose-escalation, and dose-finding trials. Commonly used statistical methods for adaptive design will be introduced and compared, including the error spending approach, various methods using combinations of stagewise p-values. We will discuss implementations of adaptive trials, including interim monitoring, dynamic randomization, and analyses of adaptive trials. Practical examples using SAS and ExpDesign Studio will be provided. The FDA guidance on adaptive clinical trial designs and challenges will be discussed with recommendations. After the class, the attendees are expected to have basic knowledge to start his/her own adaptive trial design with confidence.

Brief Biography

Dr. Mark Chang, the executive director, leads the Department of Biostatistics and Data Management with 16 years of experience as a statistician in the field of clinical trials. In addition, he has over 4 years of teaching experience as assistant professor. Before joining AMAG, Chang held various positions in Millennium Pharmaceuticals, including Director of Biostatistics and Scientific Fellow. He is a co-founder of the International Society for Biopharmaceutical Statistics, an executive member of the ASA Biopharmaceutical Section, and a member of the Expert Panel for the Networks of Centres of Excellence (NCE), Canada. He is a co-chair of the Biotechnology Industry Organization (BIO) Adaptive Design Working Group and member of the PhRMA Adaptive Design and Biomarker Working Groups. Dr. Chang is an associate editor for Statistic Journals and has over 40 publications including five books. He also serves on the Editorial Boards for the Journal of Biopharmaceutical Statistics, Statistics in Biopharmaceutical Research (ASA Journal), and the Open Public Health Journal. He has been invited to serve as a co-chair on the scientific advisory and organization committees for national and international professional/academic conferences on statistics and clinical trial designs. He has edited special issues for Journal of Biopharmaceutical Statistics, discussing the FDA guidance (Draft) on adaptive designs and has been invited twice to present statistical topics to the US Food and Drug Administration. He was invited by international medical journals to write opinion papers on clinical trials. He has taught over ten statistical short courses recently. He was recently interviewed by Journalists from the Nature Group and other Scientific Journals on innovative trial designs. Dr. Chang is an adjunct professor of Boston University and an elected fellow of the American Statistical Association.

Statistical Analysis of Network Data

         Eric Kolaczyk
         Professor
         Boston University
         E-mail: kolaczyk at bu.edu

         1:00-5:00pm, April 20th, 2012 at Crosstown Center Room 460/462

Short Course Slides

Summary

Over the past decade, the study of so-called "complex networks" — that is, network-based representations of complex systems — has taken the sciences by storm. Researchers from biology to physics, from economics to mathematics, and from computer science to sociology, are more and more involved with the collection, modeling and analysis of network-indexed data. With this enthusiastic embrace of networks across the disciplines comes a multitude of statistical challenges of all sorts — many of them decidedly non-trivial. In this short course, we will cover a brief overview of the foundations common to the statistical analysis of network data across the disciplines, from a statistical perspective, in the context of topics like network summary and visualization, network sampling, network modeling and inference, and network processes. Concepts will be illustrated drawing on examples from bioinformatics, computer network traffic analysis, neuroscience, and social networks.

Brief Biography

Eric Kolaczyk is Professor of Statistics, and Director of the Program in Statistics, in the Department of Mathematics and Statistics at Boston University, where he also is an affiliated faculty member in the Program in Bioinformatics, the Program in Neuroscience, and the Division of Systems Engineering. Before coming to Boston University, he was faculty in the Department of Statistics at the University of Chicago. In addition, he has been a visiting faculty at Harvard University, the Universite de Paris VII, and l'Ecole Nationale de la Statistique et de l'Administration Economique (ENSAE) in Paris.

Prof. Kolaczyk's main research interests currently revolve around the statistical analysis of network-indexed data, and include both the development of basic methodology and inter-disciplinary work with collaborators in bioinformatics, computer science, geography, neuroscience, and sociology. Besides various research articles on these topics, he has also authored a book in this area — Statistical Analysis of Network Data: Methods and Models (Springer, 2009). He has given various short courses on material from his book in recent years, including for the Center for Disease Control (CDC) and the Statistical and Applied Mathematical Sciences Institute (SAMSI) in the US as well as similar venues in Belgium, England, and France. Prior to his working in the area of networks, Prof. Kolaczyk spent a decade working on statistical multi-scale modeling.

Prof. Kolaczyk has served as associate editor on several journals, including currently the Journal of the American Statistical Association. He has also served as co-organizer for workshops focused on networks and network data. He is an elected fellow of the American Statistical Association (ASA), an elected senior member of the Institute for Electrical and Electronics Engineers (IEEE), and an elected member of the International Statistical Institute (ISI).

Stochastic Modeling of Limit Order Books

         Rama Cont
         Professor
         Columbia University
         E-mail: rama.cont at columbia.edu

         1:00-5:00pm, April 20th, 2012 at Crosstown Center Room 460/462

Summary

An increasing proportion of financial transactions take place in electronic markets where buy and sell orders submitted by market participants are centralized in a limit order book and executed according to precise time and price priority rules. The availability of (TeraBytes of) high-frequency data on limit order books offer a fascinating glimpse into the dynamics prices, supply and demand in financial markets and pose interesting challenges in terms of statistical modeling, both for market participants and for those — regulators and economists — who seek to understand the consequences of high frequency trading.

This course will serve as an introduction to the statistical modeling of limit order books: after describing the nature of the data and the time scales involved and reviewing some of the statistical properties of limit order books, we will argue that a limit order book has a natural description in terms of a spatial point process or queueing system, and provide various example of point process models proposed in the recent literature.

Applications of such models involve time scales ranging from the millisecond (interval between orders) and the day (time needed to liquidate a large batch of shares). In the second part of the lectures, we show how functional limit theorems may be used as a useful tool to link high-frequency behavior of order flow to features such as price volatility and autocorrelation of price movements at lower frequencies. We will use fluid limits and functional central limit theorems to shows that, in liquid markets where orders arrive with high frequency, the dynamics of buy and sell queues may be approximated by a Markovian jump-diffusion process. This approximation provides an analytically tractable description of the dynamics of the order book and the market price and yields a quantitative link between statistical properties of the price process and properties of the order flow.

Finally, we will sketch some open problems and challenges posed by large high-frequency data sets and discuss the potential for statistical learning methods for studying these issues.

References

1. Rama Cont, Sasha Stoikov and Rishi Talreja (2010) A stochastic model for order book dynamics, Operations Research, Volume 58, No. 3, 549-563. Abstract.
2. Rama Cont and Adrien de Larrard (2010) Price dynamics in a Markovian limit order market. Abstract.
3. Rama Cont (2011) Statistical Modeling of High Frequency Financial Data: Facts, Models and Challenges, IEEE Signal Processing, Vol 28, No 5, 16-25. Abstract.
4. Rama Cont and Adrien de Larrard (2011) Order book dynamics in liquid markets: limit theorems and diffusion approximations. Abstract.
5. Rama Cont and Adrien de Larrard (2011) Price dynamics in limit order markets: linking volatility and order flow, Working Paper.
6. Allan Andersen, Rama Cont, and Ekaterina Vinkovskaya (2011) A self-exciting point process model for the high-frequency dynamics of limit order books, Working Paper, Columbia University.
7. Kirilenko, Andrei A., Kyle, Albert S., Samadi, Mehrdad and Tuzun, Tugkan, The Flash Crash: The Impact of High Frequency Trading on an Electronic Market (May 2011). Abstract.

Brief Biography

Rama CONT is Associate professor at Columbia University (New York), director of the Columbia Center for Financial Engineering and CNRS Research Scientist at Laboratoire de probabilites (Universite de Paris VI). His research deals with stochastic analysis and stochastic modeling of financial risks, with a focus on the modelling of extreme risks — market discontinuities, systemic risk and endogenous risk — in financial markets. He was awarded the Louis Bachelier Prize by the French Academy of Sciences in 2010 for his research on mathematical modeling in finance. He is co-author of Financial Modeling with Jump Processes (CRC Press, 2003) and the Editor-in-chief of the Encyclopedia of Quantitative Finance (Wiley, 2010) and has served as a consultant to numerous financial institutions and regulatory bodies in Europe and the US.