Bruce W. Schmeiser

Learn More
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Institute of(More)
Batching is a commonly used method for calcuLating confidence intervals on the mean of a sequence of correlated observations arising from a simulation experiment. Several recent papers have considered the effect of using too many batches. The use of too many batches fails to satisfy assumptions of normaLity and/or independence, resulting in incorrect(More)
Nonoverlapping batch means (NOLBM) is a we11-known approach For estimating the variance of the sample mean. In this paper we consider an overlapping batch means (OLBM) estimator that, based on the same assumptions and batch size as NOLBM, has essentially the same mean and only 2/3 the asymptotic variance of NOLBM. Confidence interval procedures for the mean(More)
Tukey's lambda distribution is generalized to provide an algorithm for generating values of unimodal asymmetric random variables. This algorithm has the same advantages as the symmetric random variable generator previously given by the authors, except that the addition of another parameter complicates the problem of finding the parameter values to fit a(More)
Neural network (NN) based modeling often requires trying multiple networks with different architectures and training parameters in order to achieve an acceptable model accuracy. Typically, only one of the trained networks is selected as "best" and the rest are discarded. The authors propose using optimal linear combinations (OLC's) of the corresponding(More)
In this tutorial we consider three methods for selecting the best of a set of competings yst ems: interactive analysis, ranking and selection, and multiple comparisons. We describe each method; discuss assumptions, implement ation aspects, advantages, and disadvantages; and demonstrate the use of each method with an airline-reservation-system simulation(More)
A method for generating values of continuous symmetric random variables that is relatively fast, requires essentially no computer memory, and is easy to use is developed. The method, which uses a uniform zero-one random number source, is based on the inverse function of the lambda distribution of Tukey. Since it approximates many of the continuous(More)
The stochastic root-finding problem is to find the root of the equation g(z) = y, where g(z) can be estimated. There are many applications, including continuous and convex stochastic optimization, which is the problem of finding the zero of the gradient function. We propose a family of retrospective approximation algorithms that numerically solve a sequence(More)