Document Type
Lecture
Publication Date
3-19-2004
Abstract
Statistics has grown out of the need for modeling and analysis in various interdisciplinary fields. Nurtured by the advent of information (and bio-) technology, statistical learning seems to have stolen the limelight, depriving statistical methodology of its due impact, as well as, raising qualms about the basic role of statistical reasoning in applied fields. In practice, usually statistical models are not simple, and are subject to complex constraints. There is a genuine and growing need for statistical inference to cope with a constrained environment, albeit there may not be generally any optimal decision theoretic framework for constrained inference, and the path is marred by many impasses. A considerable amount of effort has been spent in developing applicable methodology to deal with such constrained statistical inference in simplistic situations, in the hope that this would pave the way for more complex models arising in modern applications. Along with an overview of the general subject matter, some specific problems are illustrated with appropriate data models.
Relational Format
presentation
Recommended Citation
Sen, Pranab K., "Constrained Inference in Statistical Practice" (2004). Probability & Statistics Seminar. 61.
https://egrove.olemiss.edu/math_statistics/61
