Workshop History

The International Workshop in Sequential Methodologies (IWSM) is a biannual conference dating back to 2007 that seeks to bring together statisticians, at all stages of their careers, from academia as well as from industry, to present, discuss, and explore the latest developments, challenges and opportunities in theoretical and applied sequential statistical methods. The Workshop covers all aspects of sequential statistics, from optimal stopping, sequential estimation and hypothesis testing, and quickest change-point detection, to applications in information theory, quality control, clinical trials, finance, and cybersecurity - to name a few.

A brief account of the Workshop history is as follows:

  1. The 1st IWSM was held in 2007 at Auburn University, Auburn, Alabama, USA;
  2. The 2nd IWSM was held in 2009 at the University of Technology of Troyes, Troyes, France;
  3. The 3rd IWSM was held in 2011 at Stanford University, Stanford, California, USA;
  4. The 4th IWSM was held in 2013 at the University of Georgia-Athens, Athens, Georgia, USA;
  5. The 5th IWSM was held in 2015 at Columbia University, New York, New York, USA;
  6. The 6th IWSM was held in 2017 at the University of Rouen, Rouen, Normandy, France.

Historically, sequential statistical methods originated within the Statistical Research Group of Columbia University in the 1940s in a project supported by the Bureau of Ordnance, Navy Department. One of the key contributors on the project was Abraham Wald who essentially set forth the entire field. The salient feature of the sequential approach to statistical inference is that the number of observations required by a sequential procedure is not fixed in advance of the experiment, but rather is determined as the experiment goes on. Consequently, sequential procedures can be developed that require, on average, a substantially smaller number of observations than equally reliable tests based on a predetermined number of observations. Because of this remarkable property of the sequential approach, its great potential and far-reaching consequences in development work on military and naval equipment were immediately recognized, and the early efforts in the field was classified “Restricted” within the meaning of the Espionage Act. The restricted classification was lifted in 1945. Since then, the field has undergone an enormous expansion in depth as well as in breadth, both on the theoretical side and on the applied side, too. Today, the need for efficient algorithms for online estimation, hypothesis testing and change-point detection arises in such areas as communications, seismology, signal process, target detection & tracking, and finance.