CONVEXITY PROPERTIES OF PROBABILITY AND QUANTILE FUNCTIONS IN OPTIMIZATION PROBLEMS

Authors
Citation
Ys. Kan et Ai. Kibzun, CONVEXITY PROPERTIES OF PROBABILITY AND QUANTILE FUNCTIONS IN OPTIMIZATION PROBLEMS, Automation and remote control, 57(3), 1996, pp. 368-383
Citations number
30
Categorie Soggetti
Controlo Theory & Cybernetics","Computer Application, Chemistry & Engineering","Instument & Instrumentation","Robotics & Automatic Control
ISSN journal
00051179
Volume
57
Issue
3
Year of publication
1996
Part
1
Pages
368 - 383
Database
ISI
SICI code
0005-1179(1996)57:3<368:CPOPAQ>2.0.ZU;2-3
Abstract
A survey is given of investigations on the convexity properties of pro bability cord quantile functions in models of stochastic programming a nd optimal control. Some conceptual contradictions of various approach es to the problem of the probability function convexity are treated. T he notions of partially g-concave functions and: probability measures g-concave relative to a convex set are introduced to obviate these con tradictions. A theorem on the convexity and g-convexity of a quantile function is proved.