A subset selection procedure under input parameter uncertainty

Corlu, Canan G.; Biller, Bahar

doi:10.1109/wsc.2013.6721442

Cited by 27 publications

(11 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although input uncertainty quantification for point estimation has been an active area of research, little has been done on the implications of input uncertainty for other analysis problems. An exception is Corlu and Biller (2013) who consider a subset selection problem in the presence of input uncertainty. We expect that many "solved" problems in stochastic simulation output analysis will be revisited taking input uncertainty into account.…”

Section: Discussionmentioning

confidence: 99%

Advanced tutorial: Input uncertainty quantification

Song

Nelson

Pegden

2014

Proceedings of the Winter Simulation Conference 2014

View full text Add to dashboard Cite

Input uncertainty" refers to the (often unmeasured) effect of not knowing the true, correct distributions of the basic stochastic processes that drive the simulation. These include, for instance, interarrival-time and service-time distributions in queueing models; bed-occupancy distributions in health care models; distributions for the values of underlying assets in financial models; and time-to-failure and time-to-repair distributions in reliability models. When the input distributions are obtained by fitting to observed realworld data, then it is possible to quantify the impact of input uncertainty on the output results. In this tutorial we carefully define input uncertainty, describe various proposals for measuring it, contrast input uncertainty with input sensitivity, and provide and illustrate a practical approach for quantifying overall input uncertainty and the relative contribution of each input model to overall input uncertainty. 162978-1-4799-7486-3/14/$31.00 ©2014 IEEE

show abstract

Section: Discussionmentioning

confidence: 99%

Advanced tutorial: Input uncertainty quantification

Song

Nelson

Pegden

2014

Proceedings of the Winter Simulation Conference 2014

View full text Add to dashboard Cite

show abstract

“…A common objective is to identify all systems whose means are within a specified distance from the mean of the best system. Corlu and Biller (2013) consider this problem while using historical data to quantify input uncertainty associated with each system. Many algorithms exist for performing multiple comparisons with a standard to determine the best system, where special consideration is given to some standard system.…”

Section: Literature Reviewmentioning

confidence: 99%

Multiple comparisons with a standard using false discovery rates

Singham¹,

Szechtman²

2016

2016 Winter Simulation Conference (WSC)

View full text Add to dashboard Cite

Multiple comparisons with a standard using false discovery rates Downloaded from NPS Archive: CalhounProceedings of the 2016 Winter Simulation Conference T. M. K. Roeder, P. I. Frazier, R. Szechtman, E. Zhou, T. Huschka, and S. E. Chick, eds. MULTIPLE COMPARISONS WITH A STANDARD USING FALSE DISCOVERY RATES ABSTRACTWe introduce a new framework for performing multiple comparisons with a standard when simulation models are available to estimate the performance of many different systems. In this setting, a large proportion of the systems have mean performance from some known null distribution, and the goal is to select alternative systems whose means are different from that of the null distribution. We employ empirical Bayes ideas to achieve a bound on the false discovery rate (proportion of selected systems from the null distribution) and a desired probability an alternate type system is selected.

show abstract

“…However, they are important building blocks for studying simulation optimization under input uncertainty. Recently, Corlu and Biller (2013) investigated a ranking-and-selection problem and develops a subset selection procedure by accounting for parameter uncertainty in the input distribution; to evaluate ranking-and-selection procedures, Waeber, Frazier, and Henderson (2010) proposed a performance analysis process that takes into account three layers of risk: the loss of decision, the configuration-specific risk (which is similar to the risk associated with input distribution in our context), and the overall risk. On the other hand, distributionally robust optimization (ORO) was first introduced by Scarf, Arrow, and Karlin (1958) in an inventory control problem and provides a nice framework for stochastic optimization under input uncertainty.…”

Section: Literature Reviewmentioning

confidence: 99%

Simulation optimization when facing input uncertainty

Zhou¹,

Xie²

2015

2015 Winter Simulation Conference (WSC)

View full text Add to dashboard Cite

Stochastic simulation is driven by the input model, which is a collection of distributions that model the randomness in the system. The input model is often constructed from data, and hence input uncertainty arises due to the finiteness of data. Simulation optimization has been mostly studied under the assumption of a known input model, without accounting for input uncertainty. We propose a new framework to study simulation optimization under input uncertainty, with the goal to balance the trade-off between optimizing under the estimated input model and hedging against the risk brought by input uncertainty. A simple numerical example illustrates different formulations under the new framework, compared with the usual formulation for simulation optimization.978-1-4673-9743-8/15/$31.00 ©2015 IEEE

show abstract

A subset selection procedure under input parameter uncertainty

Cited by 27 publications

References 22 publications

Advanced tutorial: Input uncertainty quantification

Advanced tutorial: Input uncertainty quantification

Multiple comparisons with a standard using false discovery rates

Simulation optimization when facing input uncertainty

Contact Info

Product

Resources

About