Professor Emeritus Ishay Weissman

General Information

Prof. Weissman spent five years as a lecturer at the Department of Statistics of Tel-Aviv University (1971--1976) before joining the Technion. During his tenure at the Technion, he was a visiting professor for one year each at Iowa State University (1979--1980), at Cornell University (1980--1981), at the University of California, Davis (1986--1987) and at the University of Pennsylvania (1993--1994). He also carried out research for short periods at the Universities of British Columbia (Canada), Erasmus (Holland), Surrey (UK), North Carolina (U.S.A.), New South Wales (Australia) and Australian National University. Prof. Weissman is a member of the Institute of Mathematical Statistics and the Israeli Statistical Association.

Research Summary

Prof. Weissman's research is mainly in the field of extreme value theory. The significance of this field is best described as follows: "It appears that we live in an age of disasters: the Mississippi and the Missouri rivers flood millions of acres, earthquakes hit Tokyo and California, airplanes crash due to mechanical failure, and powerful windstorms cause increasingly costly damage. While these may seem to be unexpected phenomena to the man in the street, they are actually happening according to well defined rules of science known as extreme value theory. For many phenomena records must be broken in the future, so if a design is based on the worst case of the past then we are not really prepared for the future. Materials will fail due to fatigue: even if the body of an aircraft looks fine to the naked eye, it might suddenly fail if the aircraft has been in operation over an extended period of time. Extreme value theory has by now penetrated the social sciences, the medical profession, economics and even astronomy" (Janos Galambos, National Institute of Standards and Technology, 1994, Vol. 99).

Our research is concerned with limit distributions of extremes, when the sample size grows indefinitely. More specifically, we study possible models to describe extreme phenomena for sequences of independent and identically distributed observations, dependent stationary sequences as well as nonstationary ones. Estimation methods are proposed and their properties are studied.

Current Research Projects

Extreme values and the extremal index (a BSF grant with R.L. Smith)

Fitting a Mejzler distribution to extreme value   data

On the estimation of extreme tail probabilities via bootstrapping (with P. Hall)

Selected Publications