Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Istanbul Ticaret Uni...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
versions View all 2 versions
addClaim

Variance reduction via importance sampling

Authors: YÖN, Semih; GOLDSMAN, Dave;

Variance reduction via importance sampling

Abstract

Variability always occurs to be the most frighteningphenomena in implementation of various kinds of experiments. We desire to control variability and decrease the variance of experiments in order to be aware of the accuracy of the constructed models and consequently supply reliable results. Importance Sampling, also called Biased Sampling is one of the variance reduction techniques especially used in Monte Carlo Methods. This study includes a research to gather the appropriate importance sampling density which gives the lowest variance. We illustrate the importance sampling method on an M/M/1 queuing problem involving a limited waiting capacity of 50 of buffer size and solve it with an efficient C coded simulation program. We first execute naïve simulation, afterwards we carried out importance sampling method and supplied meaningful decrease in the estimated variance of the case which queue length ever exceeds buffer size. By this way, one can calculate any expectation that cannot be calculated by analytically. Numerical results indicate that longer tailed proposal distributions provide much more meaningful decrease.

Değişkenlik veya rassal sayılara bağlı hata çeşitli deneylerde ortaya çıkan en korkutucu problemlerdendir. Gerçeğe uygun modeller kurup bunlardan güvenilir sonuçlar elde etmek istenir. Bunun için Monte Carlo uygulamalarında tahmini varyansı azaltan Taraflı Örnekleme (Importance Sampling) yöntemi kullanılabilir. Bu çalışmada en az varyansı veren dağılımlar bulunmaya çalışılmıştır. Bunun için basit bir M/M/1 kuyruk sistemi benzetim modellemesi ile analiz edilmiş ve 50 birimlik bir ön tamponun dolup aşılma olasılığı bulunmaya çalışılmıştır. Önce basit Monte Carlo benzetim modeli daha sonar Taraflı Örnekleme benzetim modeli kullanılarak sonuçlar alınmıştır ve sayısal sonuçlar daha uzun kuyruğa sahip dağılımların daha olumlu sonuç verdiğini göstermiştir.

Country
Turkey
Related Organizations
Keywords

M/M/1 Queue, M/M/1 Kuyruk Modeli, Monte Carlo Simulation, Varyans Azaltma, Taraflı Örnekleme, Varyans Azaltma;Taraflı Örnekleme;Monte Carlo Benzetim;M/M/1 Kuyruk Modeli, Importance Sampling, Variance Reduction, Monte Carlo Benzetim

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green