Filters








1,332 Hits in 10.4 sec

Hypothesis Tests for Bernoulli Experiments: Ordering the Sample Space by Bayes Factors and Using Adaptive Significance Levels for Decisions

Carlos Pereira, Eduardo Nakano, Victor Fossaluza, Luís Esteves, Mark Gannon, Adriano Polpo
2017 Entropy  
The main objective of this paper is to find the relation between the adaptive significance level presented here and the sample size.  ...  Bayesian (Bayes factor) hypothesis tests do.  ...  Acknowledgments: The first and sixth authors are grateful to the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for financial support.  ... 
doi:10.3390/e19120696 fatcat:7i5ohw4xtvecpn7jest3gjajfa

Sequential Design of Experiments via Linear Programming [article]

Sudipto Guha, Kamesh Munagala
2013 arXiv   pre-print
We provide the first polynomial time constant-factor approximation algorithm for this class of problems.  ...  Sequentiality is a well-studied concept in decision theory, and is very desirable in domains where multiple explorations can be conducted in parallel, for instance, in the sensor network context.  ...  Acknowledgment: We would like to thank Jen Burge, Vincent Conitzer, Ashish Goel, Ronald Parr, and Fernando Pereira for helpful discussions.  ... 
arXiv:0805.2630v2 fatcat:xvsuexrkbja3jnuopfvxbck5t4

Adaptive Experiments and a Rigorous Framework for Type I Error Verification and Computational Experiment Design [article]

Michael Sklar
2022 arXiv   pre-print
This PhD thesis covers breakthroughs in several areas of adaptive experiment design: (i) (Chapter 2) Novel clinical trial designs and statistical methods in the era of precision medicine.  ...  (iii) (Chapter 4) Bandit and covariate processes, with finite and non-denumerable set of arms. (iv) (Chapter 5) A rigorous framework for simulation-based verification of adaptive design properties.  ...  Subject to typical caveats on prior selection and accurate posterior sampling, posterior inference can yield Bayes factors for testing, credible intervals for treatment effects, and decision analysis for  ... 
arXiv:2205.09369v1 fatcat:pmniftjxlffctmvlu7bp7wvehe

Individual vs. group decision-making: an experiment on dynamic choice under risk and ambiguity

Enrica Carbone, Konstantinos Georgalos, Gerardo Infante
2019 Theory and Decision  
We argue that the results might be driven by differences in the levels of ambiguity and risk attitudes between individuals and groups, extending the frequently observed pattern of groups behaving closer  ...  Using a consumption/saving laboratory experiment, we investigate behaviour in four treatments: (1) individual choice under risk; (2) group choice under risk; (3) individual choice under ambiguity and (  ...  All reported p-values were generated using ranksum MWW tests for independent samples for the group treatments and signed-rank MWW tests for the individual ones.  ... 
doi:10.1007/s11238-019-09694-8 fatcat:vzoqxxwyanbmjaublrlxnr2pmi

Why optional stopping can be a problem for Bayesians

Rianne de Heide, Peter D. Grünwald
2020 Psychonomic Bulletin & Review  
priors—which means, in most practical applications of Bayes factor hypothesis testing—resilience to optional stopping can break down.  ...  By slightly varying and extending Rouder's (Psychonomic Bulletin & Review21(2), 301–308, 2014) experiments, we illustrate that, as soon as the parameters of interest are equipped with default or pragmatic  ...  Rouder and Wagenmakers both noticed significant unclarities and errors, which prompted us to do a substantial rewrite of the article.  ... 
doi:10.3758/s13423-020-01803-x pmid:33210222 fatcat:ap5aredkhrgrfaggmbnybll33y

On evaluating stream learning algorithms

João Gama, Raquel Sebastião, Pedro Pereira Rodrigues
2012 Machine Learning  
It is also worthwhile to use the proposed methods for hypothesis testing and for change detection.  ...  These experiments point out that the use of forgetting mechanisms (sliding windows or fading factors) are required for fast and efficient change detection.  ...  The dotted line is the threshold for a significance level of 99 %. For different fading factors we got different results about the significance of the differences techniques.  ... 
doi:10.1007/s10994-012-5320-9 fatcat:g6aip6id7vfw3fbeh7cne23gmu

The Bayes/Non-Bayes Compromise: A Brief Review

I. J. Good
1992 Journal of the American Statistical Association  
Various compromises that have occurred between Bayesian and nowBayesian methods are reviewed. (A citation is provided that discusses the inevitability of compromises within the Bayesian approach.)  ...  One example deals with the masses of elementary particles, but no knowledge of physics will be assumed.  ...  This Bernoulli sampling is the binomial case of testing a multinomial for equiprobability.  ... 
doi:10.1080/01621459.1992.10475256 fatcat:iektayon4zdajdqmjsisp2t3hi

The Bayes/Non-Bayes Compromise: A Brief Review

I. J. Good
1992 Journal of the American Statistical Association  
Various compromises that have occurred between Bayesian and nowBayesian methods are reviewed. (A citation is provided that discusses the inevitability of compromises within the Bayesian approach.)  ...  One example deals with the masses of elementary particles, but no knowledge of physics will be assumed.  ...  This Bernoulli sampling is the binomial case of testing a multinomial for equiprobability.  ... 
doi:10.2307/2290192 fatcat:5dpmc3r3nng7ld27aztfsuaake

Bayesian Performance Comparison of Text Classifiers

Dell Zhang, Jun Wang, Emine Yilmaz, Xiaoling Wang, Yuxin Zhou
2016 Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16  
In the area of text classification, since the publication of Yang and Liu's seminal SIGIR-1999 paper, it has become a standard practice for researchers to apply nullhypothesis significance testing (NHST  ...  However, such a frequentist approach has a number of inherent deficiencies and limitations, e.g., the inability to accept the null hypothesis (that the two classifiers perform equally well), the difficulty  ...  In order to calculate the Bayes factor using the SD method (see Section 3.2.1), we approximate the posterior density Pr[δ = 0|M1, D] and the prior density Pr[δ = 0|M1] by fitting a smooth function to the  ... 
doi:10.1145/2911451.2911547 dblp:conf/sigir/ZhangWYWZ16 fatcat:h642m4qr2zg3befxu2himexj2q

Application of an Adaptive Incremental Classifier for Streaming Data
스트리밍 데이터에 대한 적응적 점층적 분류기의 적용

Cheong Hee Park
2016 Journal of KIISE  
A hypothesis test is then performed for concept drift detection. Based on the estimated p-value, the weight of outdated data is set automatically in updating the classifier.  ...  A distribution, representing the performance pattern of a classifier, is constructed by utilizing the distance between the confidence score of a classifier and a class indicator vector.  ...  The evaluation methodology used in our experiments is the Interleaved Test-and-Train approach. Each data sample is first used for testing, and then it is used to train the model.  ... 
doi:10.5626/jok.2016.43.12.1396 fatcat:rqpjfbrshnarxpfio63a4wa3wm

SReach: A Bounded Model Checker for Stochastic Hybrid Systems [article]

Qinsi Wang, Paolo Zuliani, Soonho Kong, Sicun Gao, Edmund M. Clarke
2014 arXiv   pre-print
Our approach encodes stochastic information by using random variables, and combines the randomized sampling, a δ-complete decision procedure, and statistical tests.  ...  The statistical tests adapted guarantee arbitrary small error bounds between probabilities estimated by SReach and real ones.  ...  The use of Bayes factors is a Bayesian alternative to classical hypothesis testing. It is based on the Bayes theorem.  ... 
arXiv:1404.7206v2 fatcat:w2ksrikju5hibc2qqmuinsekxm

Sequential Bayes-Optimal Policies for Multiple Comparisons with a Known Standard

Jing Xie, Peter I. Frazier
2013 Operations Research  
unknown sampling variance, Bernoulli, and Poisson.  ...  The most straightforward approach for allocating sampling effort, and the approach most commonly employed by practitioners, is to simulate each system an equal number of times.  ...  The authors would like to thank Matthew S. Maxwell and Shane G. Henderson for the use of their ambulance simulation software, and their help in using it.  ... 
doi:10.1287/opre.2013.1207 fatcat:46n3cc5glzbsjf5vd7ivzs5xja

Bayesian techniques for analyzing group differences in the Iowa Gambling Task: A case study of intuitive and deliberate decision-makers

Helen Steingroever, Thorsten Pachur, Martin Šmíra, Michael D. Lee
2017 Psychonomic Bulletin & Review  
Most commonly, IGT behavior is analyzed using frequentist tests to compare performance across groups, and to compare inferred parameters of cognitive models developed for the IGT.  ...  The three model-based methods involve Bayesian hierarchical parameter estimation, Bayes factor model comparison, and Bayesian latent-mixture modeling.  ...  From such non-significant results of frequentist tests, one can only conclude that the null hypothesis cannot be rejected.  ... 
doi:10.3758/s13423-017-1331-7 pmid:28685273 pmcid:PMC5990582 fatcat:rtkwnyepsre6xlcuip2g7v4wdy

Restricted most powerful Bayesian tests for linear models

Scott D. Goddard, Valen E. Johnson
2016 Scandinavian Journal of Statistics  
I illustrate the use of RMPBTs in the special cases of ANOVA and one-and two-sample t-tests.  ...  This correspondence leads to the definition of default Bayes factors for many common tests of linear hypotheses.  ...  The main function, bas, can either search the model space exhaustively when there are less than 25 covariates or use adaptive sampling without replacement for larger model spaces.  ... 
doi:10.1111/sjos.12235 fatcat:ybsn7hl3lrg4lggval3slo27sq

Prioritizing tests for software fault diagnosis

Alberto Gonzalez-Sanchez, Éric Piel, Rui Abreu, Hans-Gerhard Gross, Arjan J. C. van Gemund
2011 Software, Practice & Experience  
However, test suites that are prioritized for failure detection can reduce the amount of useful information for fault localization.  ...  During regression testing, test prioritization techniques select test cases that maximize the confidence on the correctness of the system when the resources for quality assurance (QA) are limited.  ...  We also thank our anonymous reviewers for their helpful comments and insights, which greatly improved the quality of this paper.  ... 
doi:10.1002/spe.1065 fatcat:7v2ovemwo5hu7fnafsyzd7b2lm
« Previous Showing results 1 — 15 out of 1,332 results