Search

IIASA celebrated its twentieth anniversary on May 12-13 with its fourth general conference, IIASA '92: An International Conference on the Challenges to Systems Analysis in the Nineties and Beyond. The conference focused on the relations between environment and development and on studies that integrate the methods and findings of several disciplines. The role of systems analysis, a method especially suited to taking account of the linkages between phenomena and of the hierarchical organization of the natural and social world, was also assessed, taking account of the implications this has for IIASA's research approach and activities. This paper is one of six IIASA Collaborative Papers published as part of the report on the conference, an earlier instalment of which was Science and Sustainability, published in 1992. The term "identification" came into use by economists in the late 1920s, but the general idea has existed at least as long as the use of mathematics in science, and is applicable to natural as well as social phenomena. Identification is finding the underlying structure that generated the observed data. In fact we never find the structure itself, but at best a model that is uniquely capable of doing what the structure does. Usually identification is sought by solving for the values of parameters in a given set of equations -- often linear equations. Professor Deistler would broaden the search beyond finding the right coefficients in a set of linear equations; his method permits the use of intuition as well as fitting to find the most likely model.