Optimality of Certain Asymmetrical Experimental Designs
نویسندگان
چکیده
منابع مشابه
Equivalence theorem for Schur optimality of experimental designs
An experimental design is said to be Schur optimal, if it is optimal with respect to the class of all Schur isotonic criteria, which includesKiefer’s criteria of p-optimality, distance optimality criteria andmany others. In the paper we formulate an easily verifiable necessary and sufficient condition for Schur optimality in the set of all approximate designs of a linear regression experiment w...
متن کاملOptimality Aspects of Designs
1. Introduction Careful experimentation is an essential part of any problem of decision-making. An experiment is a device or means of getting an answer to the problem under consideration. In any sphere of scientific research, the general procedure that the experimenter has to adopt is to formulate the hypothesis according to the problem whose answer is to be sought for and then to verify the fo...
متن کاملCertain Hadamard Designs
(2.1) F = C+C0 + C If A, B are two aggregates of elements of F we shall denote by AB the aggregate formed by adding each element of A to every element of B. We shall also denote the aggregate obtained by taking A a times by a A. Then we have the following Lemma 1. If p'=l (mod 4), then pl1 C0Ce = ——(C. + o, 4 , pl 1 p'-S pl-i (2.2) 0-, = ^-—— C + Í-—— C. + --C., 2 4 4 ¿i _ i ¿i _ i ¿i _ s c. = ...
متن کاملOptimality of Certain Channel Aware Scheduling Policies
With an abstraction of serving rate-adaptive sources on a broadcast-type wireless channel as a utility maximization problem, it is shown how one can design many intuitive online scheduling policies based upon the feedback that one obtains at the scheduler. Using a stochastic approximation argument it is then shown that the constructed algorithms converge to optimal solutions of the utility maxi...
متن کاملAsymptotic Optimality of Certain Multi - Alternativesequential Tests
It is known that certain combinations of one-sided sequential probability ratio tests are asymptotically optimal (relative to the expected sample size) for problems involving a nite number of possible distributions when probabilities of errors tend to zero and observations are independent and identically distributed according to one of the underlying distributions. The objective of this paper i...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: The Annals of Statistics
سال: 1978
ISSN: 0090-5364
DOI: 10.1214/aos/1176344371