Lorden, Gary (1977) Nearly-optimal sequential tests for finitely many parameter values. Annals of Statistics, 5 (1). pp. 1-21. ISSN 0090-5364 http://resolver.caltech.edu/CaltechAUTHORS:LORas77
See Usage Policy.
Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:LORas77
Combinations of one-sided sequential probability ratio tests (SPRT's) are shown to be "nearly optimal" for problems involving a finite number of possible underlying distributions. Subject to error probability constraints, expected sample sizes (or weighted averages of them) are minimized to within o(1) asymptotically. For sequential decision problems, simple explicit procedures are proposed which "do exactly what a Bayes solution would do" with probability approaching one as the cost per observation, c, goes to zero. Exact computations for a binomial testing problem show that efficiencies of about 97% are obtained in some "small-sample" cases.
|Additional Information:||© 1977 Institute of Mathematical Statistics. Received December 1974; revised June 1976. Research supported by the National Science Foundation under Grant GP-37819.The author is grateful to David Siegmund and Michael Klass for stimulating and helpful conversations, to the Associate Editor for a careful reading of the manuscript and useful suggestions, and especially to Jack Kiefer for his enthusiastic and inspiring guidance of the author's thesis research, of which the present work is an outgrowth.|
|Subject Keywords:||Sequential probability ratio test, Bayes solution, asymptotic optimality|
|Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Tony Diaz|
|Deposited On:||25 Jan 2007|
|Last Modified:||26 Dec 2012 09:30|
Repository Staff Only: item control page