Allocation of Observations in Ranking and Selection With Unequal Variances

A Multivariate Solution of the Multivariate Ranking and Selection Problem

“…PROOF. This follows from the proof of Theorem (5.1) of Dudewicz and Bishop (1979), where we find that the limiting (as n. distribution of (Xl'. 7k) is the same as the distribution of (YI, "",Yk ) .…”

“…This can be used to achieve (2.8) with a smaller total sample size However, since (5.1) is the type of probability requirement one would usually desire in practice, it is desirable to seek a satisfying (5.1 in a broader class than the class of singlestage procedures. Dudewicz and Bishop (1979) have given a rtethod, called…”

Proceedings of the Twentieth Conference on the Design of Experiments in Army Research Development and Testing. Part 2. Ranking and Selection Procedures

“…Remark 11: Paulson [1964), Section 5, proposed an open-ended sequential procedure which permanently eliminates non-contending populations; his procedure is applicable in situations in which the common variance is unknown, 2.5.2 Two-stage procedures for the compýltely unknown variance case Dudewicz and Dalal [1971), and also Rinott [1974), proposed twostage procedures for the completely unknown variance case. Like (2.9), the common number of observations in the first stage for each of these procedures is arbitrary (>l)o while the total number of observations per population is a random variable.…”

“…If he wishes to guarantee (2.2) and adopts assumption (2.5c), then he cannot use a single-staee procedure (see Dudewicz [1971]) although he can use a two-stage procedure (Bechhofer, Dunnett, Sobel [19543) or a sequential procedure (Paulson [1964]); similarly, if he wishes to guarantee (2.2) and adopts assumption (2.5d), then he cannot use a single-stage procedure although he can use a two-stage procedure (Dudewicz and Dalal [1971) or Rinott [1974)). Finally, if the experimenter wishes to guarantee (2.4), and.…”

Operational decision making for medical clinics through the use of simulation and multi-attribute utility theory.

“…Another different and popular way to select the best systems is due to Dudewicz and Dalal (1975). Their method guarantees that the performance measure value of the selected differs from the optimal solution value by at most a small amount , with a probability of at least * .…”