Generalized Neyman-Pearson lemma via convex duality
Cvitanic, Jaksa ; Karatzas, Ioannis
Bernoulli, Tome 7 (2001) no. 6, p. 79-97 / Harvested from Project Euclid
We extend the classical Neyman-Pearson theory for testing composite hypotheses versus composite alternatives, using a convex duality approach, first employed by Witting. Results of Aubin and Ekeland from non-smooth convex analysis are used, along with a theorem of Komlós, in order to establish the existence of a max-min optimal test in considerable generality, and to investigate its properties. The theory is illustrated on representative examples involving Gaussian measures on Euclidean and Wiener space.
Publié le : 2001-02-14
Classification:  hypothesis testing,  Komlós theorem,  non-smooth convex analysis,  normal cones,  optimal generalized tests,  saddle-points,  stochastic games,  subdifferentials
@article{1080572340,
     author = {Cvitanic, Jaksa and Karatzas, Ioannis},
     title = {Generalized Neyman-Pearson lemma via convex duality},
     journal = {Bernoulli},
     volume = {7},
     number = {6},
     year = {2001},
     pages = { 79-97},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1080572340}
}
Cvitanic, Jaksa; Karatzas, Ioannis. Generalized Neyman-Pearson lemma via convex duality. Bernoulli, Tome 7 (2001) no. 6, pp.  79-97. http://gdmltest.u-ga.fr/item/1080572340/