Recently, Sargent and Shanthikumar [5] developed an interesting new variance reduction technique designed to exploit the stochastic structure associated with a cyclic regenerative process. Our purpose here is to study precise conditions under which the confidence intervals proposed in [5] are asymptotically valid. This analysis will provide us with the side benefit of obtaining an optimal way o...

A common class of shielding problems involves the calculation of biological dose rates from large distributed sources at localised positions. A typical example is that of calculating the dose at points around a transport cask. Sophisticated variance reduction methods must be used in Monte Carlo calculations to sample from the important parts of the source and to propagate particles towards the ...

We present results for the strangeness contribution to the nucleon, 〈N|s̄s|N〉 and to the spin of the nucleon, ∆s. By combining several variance reduction techniques for all-to-all propagators we are able to obtain gains in terms of computer time of factors of 25–30 for the disconnected loop that is needed within the calculation of ∆s, relative to the standard approach of just employing time part...

We investigate methods for variance reduction and the elimination of systematic error in a Fourier accelerated Langevin scheme for general spin models. We present results for the SU (3) × SU (3)/SU (3) model in two-dimensions that are consistent with those from multi-grid methods. We argue that the timing for the Langevin method makes it comparable to multi-grid for a given level of error.

In this paper, we propose the calibration procedure for the variance reduction of the stratified Warner’s randomized response estimators, which suggested by Hong et al. (1994) and Kim and Warde (2004), using auxiliary information at the population level. It is shown that the proposed calibration estimators are more efficient than the ordinary Warner’s estimators.

Decision tree grafting adds nodes to an existing decision tree with the objective of reducing prediction error. A new grafting algorithm is presented that considers one set of training data only for each leaf of the initial decision tree, the set of cases that fail at most one test on the path to the leaf. This new technique is demonstrated to retain the error reduction power of the original gr...

This paper discusses the need for variance reduction in simulations order to reduce time required compute a simulation. The large and complex network is commonly evaluated using large-scale Monte Carlo Unfortunately, due different sizes of network, it takes some complete However, techniques (VRT) can help solve issues. effect VRT changes behaviors simulation, particularly run To evaluate reliab...