Aplicação da Metaheurística PSO na Identificação de Pontos Influentes por meio da Função de Sensibilidade de Casos
DOI:
https://doi.org/10.5540/tema.2011.011.01.0041Abstract
Neste trabalho é aplicada a meta heurística Otimização por Enxame de Partículas (Particle Swarm Optimization -PSO) na identificação de pontos influentes. Estes pontos exercem grande influência na determinação dos coeficientes do modelo de regressão. Foi utilizada, como função objetivo, a função de sensibilidade de casos gCook(ǫ) que tem comportamento multimodal. A eficiência da metodologia proposta foi testada em conjuntos de dados simulados. Os resultados obtidos mostram que esta metodologia apresenta soluções satisfatórias na identificação de pontos influentes.References
[1] R.A. Atkinson, Masking unmasked, Biometrika, 73 (1986), 533–541.
D. Belsley, E. Kuh, R. Welsch, “Regression Diagnostics”, John Wiley, New York, 1980.
E. Biazi, “Some Aspects of Influence Analysis and a New Approach”, Tese de doutorado, University of Warwick, 1996.
N. Billor, R. Loynes, Local influence: a new approach, Communs Statist., 22 (1993), 1595–1661.
R. Cook, Detection of influential observations in linear regression, Technometrics, 19 (1977), 15–18.
R. Cook, S. Weisberg, “Residuals and Influence in Regression”, Chapman and Hall, 1982.
R. Cook, Assessment of local influence, J.R.Statist.Assoc., 48 (1986), 133–169.
A. Costa, E. Biazi, J. Vítor, Aplicação da metaheurística PSO na identificação de pontos influentes por meio da função de sensibilidade de casos,Anais do CNMAC, 2 (2009), 586–591.
K.D. Crawford, R.L. Wainwright, Applying genetic algorithms to outlier detection, em “Proceedings of The Sixth International Conference on Genetic Algorithms”, Morgan Kaufmann Publishers, 546–550, 1986.
F. Critchley, Discussion of leave-k-out diagnostics for time series by A.G.Bruce and R.D.Martin, J. R. Statist. Soc., 51 (1989) 407–408.
F. Critchley, R.A. Atkinson, G. Lu, E. Biazi, Influence analysis basead on the case sensitivity function, Royal Statistical Society, 63 (2001), 307–323.
R. Eberhart, P. Simpson, R. Dobbins, “Computational Intelligence PC Tools”, MA Academic Press Professional, Boston, 1996.
R. Eberhart, Y. Shi, Comparing inertia weights and constriction factors in particle swarm optimization, em “Proceedings of Congress on Evolutionary Computation”, San Diego, 84–88, 2000.
F. Hampel, The influence curve and its role in robust estimation, J. Amer. Statist. Assoc., 69, (1974), 383–393.
J. Kennedy, R. Eberhart, Particle swarm optimization, em “Proc. of the IEEE. International Conference on Neural Networks”, Piscataway, NJ, 1942–1948, 1995.
D. Peña, A new statistic for influence in linear regression, Technometrics, 47, (2005), 1–12.
R. Poli, Analysis of the publications on the applications of particle swarm optimization, J.Artif. Evol. Ap , 1,(2008), 1–10.
P.J. Rousseeauw, A.M. Leroy, “Robust Regression and Outlier Detection”, New York, Wiley, 1987.
Y. Shi, R. Eberhart, A modified particle swarm optimizer, em “Proc. of the IEEE Congress on Evolutionary Computation”, Piscataway, NJ, 69–73, 1998.
J. Tolvi, Genetic algorithms for outlier detection and variable selection in linear regression models, Soft Computing, 8 (2004), 527–533.
D. Ye, Z. Chen, A new algorithm for high-dimensional outlier detection based on constrained particle swarm intelligence, Lecture Notes in Computer Science, 5009, (2008), 516–523.
Downloads
Published
How to Cite
Issue
Section
License
Authors who publish in this journal agree to the following terms:
Authors retain copyright and grant the journal the right of first publication, with the work simultaneously licensed under the Creative Commons Attribution License that allows the sharing of the work with acknowledgment of authorship and initial publication in this journal.
Authors are authorized to assume additional contracts separately, for non-exclusive distribution of the version of the work published in this journal (eg, publish in an institutional repository or as a book chapter), with acknowledgment of authorship and initial publication in this journal.
Authors are allowed and encouraged to publish and distribute their work online (eg, in institutional repositories or on their personal page) at any point before or during the editorial process, as this can generate productive changes as well as increase impact and the citation of the published work (See The effect of open access).
This is an open access journal which means that all content is freely available without charge to the user or his/her institution. Users are allowed to read, download, copy, distribute, print, search, or link to the full texts of the articles, or use them for any other lawful purpose, without asking prior permission from the publisher or the
author. This is in accordance with the BOAI definition of open access
Intellectual Property
All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License under attribution BY.