====== CE-718: Métodos Computacionalmente Intensivos ====== Arquivos/páginas serão atualizados durante o curso. ===== Detalhes da oferta da disciplina ===== - **Período:** segundo trimestre de 2011, no programa [[http://www.cpgmne.ufpr.br|PGMNE]] (Pós Graduação em Métodos Numéricos em Engenharia) - **Matrículas e informações:** com Maristela, na secretaria do PGMNE - **Professor Responsável:** [[http://www.leg.ufpr.br/~paulojus|Paulo Justiniano Ribeiro Jr]], ([[http://www.leg.ufpr.br|LEG: Laboratório de Estatística e Geoinformação]]) - **Horários e Locais:** * As aulas serão no LEG (Laboratório de Estatística e Geoinformação) * Horário: Terças, 9:00 - 12:00 * **Atenção:** A primeira aula do curso na //terça, 31/05/2011//. - **Avaliação:** a ser definida ===== Programa da Disciplina ===== ===== Material do Curso ===== O material básico para o curso serão as seguinte notas. * {{:projetos:mci:cimnotes.pdf|Notas para o curso}} Entretanto vários materiais adicionais serão utilizados e/ou montados ao longo do curso. (ver na pagina do LEG a sessão de MCI) * {{:disciplinas:ce718:cim_0.0.1.tar.gz|Pacote com códigos e dados}} das notas de aula * [[projetos:mci|Coleção de Exemplos de Métodos Computacionalmente Intensivos]] (estes materiais foram produzidos em anos/estudos anteriores e deverão ser estudados, **corrigidos se necessário**, expandidos, discutidos, etc) ==== Materiais relacionados ==== * Os [[http://www.leg.ufpr.br/doku.php/disciplinas:ce709|materiais sobre verossimilhança e inferência]] podem ser úteis para consultas. De certa forma este curso de MCI via atacar problemas nos quais os métodos analíticos ou numéricos de inferência não são suficientes. ===== Referências Bibliográficas ===== @book{robert_introducing_2009, edition = {1}, title = {Introducing Monte Carlo Methods with R}, isbn = {9781441915757}, publisher = {Springer Verlag}, author = {Christian P. Robert and George Casella}, month = dec, year = {2009} }, @book{gamerman_markov_2006, edition = {2}, title = {Markov Chain Monte Carlo: Stochastic Simulation for Bayesian Inference, Second Edition}, isbn = {1584885874}, shorttitle = {Markov Chain Monte Carlo}, publisher = {Chapman and {Hall/CRC}}, author = {Dani Gamerman and Hedibert F. Lopes}, month = may, year = {2006} }, @book{albert_bayesian_2009, edition = {2nd ed.}, title = {Bayesian Computation with R}, isbn = {0387922970}, publisher = {Springer}, author = {Jim Albert}, month = may, year = {2009} }, @book{gelman_bayesian_2003, edition = {2}, title = {Bayesian Data Analysis, Second Edition}, isbn = {9781584883883}, publisher = {Chapman and {Hall/CRC}}, author = {Andrew Gelman and John B. Carlin and Hal S. Stern and Donald B. Rubin}, month = jul, year = {2003} }, @book{carlin_bayesian_2008, edition = {3}, title = {Bayesian Methods for Data Analysis, Third Edition}, isbn = {1584886978}, publisher = {Chapman and {Hall/CRC}}, author = {Bradley P. Carlin and Thomas A. Louis}, month = jun, year = {2008} }, @book{robert_introducing_2009, edition = {1}, title = {Introducing Monte Carlo Methods with R}, isbn = {9781441915757}, publisher = {Springer Verlag}, author = {Christian P. Robert and George Casella}, month = dec, year = {2009} }, @book{gilks_markov_1995, edition = {1}, title = {Markov Chain Monte Carlo in Practice}, isbn = {0412055511}, publisher = {Chapman and {Hall/CRC}}, author = {{W.R.} Gilks and S. Richardson and David Spiegelhalter}, month = dec, year = {1995} }, @book{robert_monte_2004, edition = {2nd}, title = {Monte Carlo Statistical Methods}, isbn = {0387212396}, publisher = {Springer}, author = {Christian Robert and George Casella}, month = jul, year = {2004} }, @book{manly_randomization_2006, edition = {3}, title = {Randomization, Bootstrap and Monte Carlo Methods in Biology, Third Edition}, isbn = {9781584885412}, publisher = {Chapman and {Hall/CRC}}, author = {Bryan {F.J.} Manly}, month = aug, year = {2006} } @book{gentle_handbook_2004, edition = {1}, title = {Handbook of Computational Statistics}, isbn = {3540404643}, publisher = {Springer}, author = {{J.E.} Gentle and Wolfgang {HSrdle}}, month = aug, year = {2004} }, @book{kroese_handbook_2011, edition = {1}, title = {Handbook of Monte Carlo Methods}, isbn = {9780470177938}, publisher = {Wiley}, author = {Dirk P. Kroese and Thomas Taimre and Zdravko I. Botev}, month = mar, year = {2011} }, @book{suess_introduction_2010, edition = {1st Edition.}, title = {Introduction to Probability Simulation and Gibbs Sampling with R}, isbn = {{038740273X}}, publisher = {Springer}, author = {Eric A. Suess and Bruce E. Trumbo}, month = jun, year = {2010} }, @book{kalos_monte_2008, edition = {2}, title = {Monte Carlo Methods}, isbn = {{352740760X}}, publisher = {{Wiley-VCH}}, author = {Malvin H. Kalos and Paula A. Whitlock}, month = nov, year = {2008} }, @book{monahan_numerical_2011, edition = {2}, title = {Numerical Methods of Statistics}, isbn = {0521139511}, publisher = {Cambridge University Press}, author = {John F. Monahan}, month = apr, year = {2011} }, @book{gentle_random_2003, edition = {2nd}, title = {Random Number Generation and Monte Carlo Methods}, isbn = {0387001786}, publisher = {Springer}, author = {James E. Gentle}, month = jun, year = {2003} } ===== Programas computacionais ===== * Programa básico do curso - [[http://www.r-project.org|The R project for Statistical Computing]]: página do programa **R** - [[http://leg.ufpr.br/~paulojus/embrapa/Rembrapa|Um material sobre o uso do R]] * Recursos auxiliares - Editor de texto \LaTeX: O [[http://www.miktex.org|MiKTeX]] disponibiliza arquivos de instalação para ambiente Windows - O [[http://sourceforge.net/projects/tinn-r|Tinn-R]] é um GUI/Editor para o ambiente [[http://www.r-project.org/|R]] sob Windows que facilita muito o uso do R neste ambiente operacional - O [[http://www.xemacs.org|Xemacs]] é uma outra opção de editor que facilita a edição de arquivos do \LaTeX e **R** - O [[http://www.toolscenter.org|TeXniccenter]] é um editor para ambiente windows que facilita a edição de documentos do \LaTeX ===== Histórico das aulas e atividades recomendadas ===== Veja aqui o [[disciplinas:ce718:historico2011|histórico das aulas]] do curso. ===== Atividades do curso ===== [[disciplinas:ce718:atividades2011|Atividades dos participantes]] ===== Espaço Aberto ===== [[disciplinas:ce718:aberto2011|Página aberta]] para edição pelos participantes do curso. ===== Links ===== == Aproximação de Laplace == * {{http://www.stats.ox.ac.uk/~steffen/teaching/bs2HT9/laplace.pdf|Laplace's Method of Integration - Ste en Lauritzen}}; * {{http://www.stanford.edu/~mch/harding-hausman-laplace.pdf|Using a Laplace Approximation to Estimate the Random Coefficients Logit Model by Non-linear Least Squares}}; * {{http://www.cemmap.ac.uk/wps/cwp0601.pdf|USING A LAPLACE APPROXIMATION TO ESTIMATE THE RANDOM COEFFICIENTS LOGIT MODEL BY NON-LINEAR LEAST SQUARES}}; * {{http://www.cs.berkeley.edu/~jordan/courses/260-spring10/lectures/lecture16.pdf|Laplace approximation review}}; * {{http://www.cs.toronto.edu/~mackay/itprnn/ps/343.344.pdf|Laplace's Method}}; * {{http://www.ece.rice.edu/~vc3/elec633/graphical_models_notes_091108.pdf|Laplace Approximation}}; * {{http://galton.uchicago.edu/~pmcc/pubs/paper26.pdf|Laplace approximation of high dimensional integrals}}; * {{http://support.sas.com/documentation/cdl/en/statug/63347/HTML/default/viewer.htm#statug_glimmix_a0000001432.htm|Maximum Likelihood Estimation Based on Laplace Approximation}}; * {{http://statmath.wu.ac.at/research/talks/resources/MultIRT.pdf|Fitting Multidimensional Latent Variable Models using an Efficient Laplace Approximation}}; * {{http://prin08.uniud.it/tl_files/prin08/upload/papers/2010_3.pdf|LAPLACE APPROXIMATION IN MEASUREMENT ERROR MODELS}}; * {{http://www.unc.edu/~vangelis/files/sglmmlapl.pdf|Asymptotic inference for Spatial GLMM using high order Laplace approximation}}; * {{http://www.jstor.org/pss/1390617}}; * {{http://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1003&context=statisticsdiss&sei-redir=1#search=%22laplace%20approximation%20integral%22|FULLY EXPONENTIAL LAPLACE APPROXIMATION EM ALGORITHM FOR NONLINEAR MIXED EFFECTS MODELS}}; * {{http://proquest.umi.com/pqdlink?Ver=1&Exp=07-02-2016&FMT=7&DID=1188875391&RQT=309&attempt=1&cfc=1|Applications of Laplace approximation for hierarchical generalized linear models in educational research}}; * {{http://people.math.aau.dk/~rw/Undervisning/Topics/Handouts/6.hand.pdf|Computation of the likelihood function for GLMMs}}; * {{http://www.ansci.wisc.edu/morota/beamer/computing.pdf|Computing: Generalized, Linear, and Mixed Models}}; * {{http://jmlr.csail.mit.edu/papers/volume12/cseke11a/cseke11a.pdf|Approximate Marginals in Latent Gaussian Models}}; * {{http://biowww.dfci.harvard.edu/~yili/spa1.pdf|Modeling Spatial Survival Data Using Semiparametric Frailty Models}}; * {{http://actuaryzhang.com/seminar/topic5_mcmc.pdf|Markov Chain Monte Carlo Methods}}; * 8-O{{http://dirk.eddelbuettel.com/blog/2011/07/05/#rcppeigen_introduction|Even faster linear model fits with R using RcppEigen}}; * {{http://dirk.eddelbuettel.com/blog/2011/07/14/#rcpp_gibbs_example|MCMC and faster Gibbs Sampling using Rcpp}}; * {{http://darrenjw.wordpress.com/2011/07/16/gibbs-sampler-in-various-languages-revisited|Gibbs sampler in various languages (revisited)}}; == Métodos Monte Carlo == * {{http://elsa.berkeley.edu/reprints/misc/understanding.pdf|Understanding the Metropolis-Hastings Algorithm}}; * {{http://www.econ.upenn.edu/~jesusfv/LectureNotes_7_MH|Metropolis-Hasting Algorithm - Jesús Fernández-Villaverde}}; * {{http://www.dme.ufrj.br/marina/MCMC.pdf| MCMC - Marina}}; * {{http://www.maths.bris.ac.uk/~manpw/teaching/folien1.pdf|Monte Carlo Methods: Lecture 1: Introduction - Nick Whiteley}}; * {{http://www.maths.bris.ac.uk/~manpw/teaching/folien2.pdf|Monte Carlo Methods: Lecture 2: Transformation and Rejection - Nick Whiteley}}; * {{http://www.maths.bris.ac.uk/~manpw/teaching/folien3.pdf|Monte Carlo Methods: Lecture 3: Importance Sampling - Nick Whiteley}}; * {{http://www.maths.bris.ac.uk/~manpw/teaching/folien45.pdf|Monte Carlo Methods: Lectures 5 & 6: The Gibbs Sampler - Nick Whiteley}}; * {{http://www.maths.bris.ac.uk/~manpw/teaching/folien6.pdf|Monte Carlo Methods: Lecture 7: The Metropolis-Hastings Algorithm - Nick Whiteley}}; * {{http://www.maths.bris.ac.uk/~manpw/teaching/folien78.pdf|Monte Carlo Methods:: Lectures 9 & 10: Combining Kernels, Convergence Diagnostics - Nick Whiteley}}; * {{http://www.maths.bris.ac.uk/~manpw/teaching/folien9.pdf|Monte Carlo Methods: Reversible Jump MCMC - Nick Whiteley}}; * {{http://www.maths.bris.ac.uk/~manpw/teaching/notes.pdf|Monte Carlo Methods - Lecture Notes - Edited by Nick Whiteley}}; * {{http://www.icmc.usp.br/~ehlers/SME0809/praticas/node18.html|Algoritmo de Metropolis-Hastings}}; * {{http://www.people.fas.harvard.edu/~plam/teaching/methods/mcmc/mcmc.pdf|MCMC Methods: Gibbs Sampling and the Metropolis-Hastings Algorithm - Patrick Lam}}; * {{http://www.maths.manchester.ac.uk/~pneal/CIS/CIS2007.html|Computationally Intensive Statistics 2010/2011}}; * {{http://www.maths.manchester.ac.uk/~pneal/statscomp.html|Statistical Computing 2010/2011}}; * {{http://www.lisa.stat.vt.edu/?q=node/1784|Bayesian Methods for Regression in R - Nels Johnson}}; == Algorítmo EM == * [[http://www.leg.ufpr.br/~paulojus/EM|Link para diversos artigos e materiais sobre EM]] * Outros em modelos não lineares: * [[http://www.jstor.org/stable/2533054|Walker]]: An EM Algorithm for Nonlinear Random Effects Models * [[http://bmsr.usc.edu/Core%20Research/dzd/6614.pdf|Wang et al.]]: Nonlinear random effects mixture models: Maximum likelihood estimation via the EM algorithm * [[http://dl.acm.org/citation.cfm?id=1225091|Wang]]: EM algorithms for nonlinear mixed effects models * [[http://fedc.wiwi.hu-berlin.de/xplore/ebooks/html/csa/node45.html|material online]]