Research Catalog

Details

Additional authors

1. Bertrand, Frédéric
2. Droesbeke, Jean-Jacques
3. Saporta, G. (Gilbert)
4. Thomas-Agnan, Christine

Description

1. xii, 355 pages : illustrations, maps; 24 cm

Subject

1. Statistics -- Methodology
2. Mathematical statistics

Contents

1. 9.3.3. Functional data / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.3.4. Intermediate conclusion / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.4. Canonical models / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.4.1. Parsimonious mixture models / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.4.2. Variable selection through regularization / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.4.3. Variable role modelling / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.4.4. Co-clustering / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.4.5. Intermediate conclusion / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.5. Future methodological challenges / Christophe Biernacki / Cathy Maugis-Rabusseau -- 10.1. Introduction / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 10.2. Model-based clustering / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 10.3. Clustering of microarray data / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 10.3.1. Microarray data / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 10.3.2. Gaussian mixture models / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 10.3.3. Application / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 10.4. Clustering of RNA-seq data / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 10.4.1. RNA-seq data / Andrea Rau / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau -- 10.4.2. Poisson mixture models / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 10.4.3. Applications / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 10.5. Conclusion / Marie-Laure Martin-Magniette / Cathy Maugis-Rabusseau / Andrea Rau -- 11.1. Functional regression models / Mathilde Mougeot -- 11.2. Data Mining using sparse approximation of the intra day load curves / Mathilde Mougeot -- 11.2.1. Choice of a generic dictionary / Mathilde Mougeot -- 11.2.2. Mining and clustering / Mathilde Mougeot -- 11.2.3. Patterns of consumption / Mathilde Mougeot -- 11.3. Sparse modeling with adaptive dictionaries / Mathilde Mougeot -- 11.4. Forecasting / Mathilde Mougeot -- 11.4.1. The experts / Mathilde Mougeot -- 11.4.2. Aggregation / Mathilde Mougeot -- 11.5. Performances & Software / Mathilde Mougeot -- 11.6. Conclusion and perspectives / Mathilde Mougeot -- 11.7. Annexes / Mathilde Mougeot.
2. 1.1. Introduction / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.2. Elements of the history of words and ideas / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.3. Modeling in astronomy / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.4. Triangulation in geodesy / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.5. The measurement of meridian arcs / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.6. A model selection tale / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.7. A new model appears / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.8. Expeditions for choosing a good model / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.9. The control of errors / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.10. A final example / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 1.11. Outline of the book / Jean-Jacques Droesbeke / Gilbert Saporta / Christine Thomas-Agnan -- 2.1. Model selection / Pascal Massart -- 2.1.1. Empirical risk minimization / Pascal Massart -- 2.1.2. The model choice paradigm / Pascal Massart -- 2.1.3. Model selection via penalization / Pascal Massart -- 2.2. Selection of linear Gaussian models / Pascal Massart -- 2.2.1. Examples of Gaussian frameworks / Pascal Massart -- 2.2.2. Some model selection problems / Pascal Massart -- 2.2.3. The least squares procedure / Pascal Massart -- 2.3. Selecting linear models / Pascal Massart -- 2.3.1. Mallows' heuristics / Pascal Massart -- 2.3.2. Schwarz's heuristics / Pascal Massart -- 2.3.3. A first model selection theorem for linear models / Pascal Massart -- 2.4. Adaptive estimation in the minimax sense / Pascal Massart -- 2.4.1. Minimax lower bounds / Pascal Massart -- 2.4.2. Adaptive properties of penalized estimators for Gaussian sequences / Pascal Massart -- 2.4.3. Adaptation with respect to ellipsoids / Pascal Massart -- 2.4.4. Adaptation with respect to arbitrary lp-bodies / Pascal Massart -- 2.5. Appendix / Pascal Massart -- 2.5.1. Functional analysis: from function spaces to sequence spaces / Pascal Massart -- 2.5.2. Gaussian processes / Pascal Massart -- 3.1. A general Theorem / Pascal Massart -- 3.2. Selecting ellipsoids and l2 regularization / Pascal Massart -- 3.2.1. Adaptation over Besov ellipsoids / Pascal Massart -- 3.2.2. A first penalization strategy / Pascal Massart -- 3.2.3. l2 regularization / Pascal Massart -- 3.3. l1 regularization / Pascal Massart -- 3.3.1. Variable selection / Pascal Massart -- 3.3.2. Selecting l1 balls and the Lasso / Pascal Massart -- 3.4. Appendix / Pascal Massart -- 3.4.1. Concentration inequalities / Pascal Massart -- 3.4.2. Information inequalities / Pascal Massart -- 3.4.3. Birge's Lemma / Pascal Massart -- 4.1. The Bayesian paradigm / Jean-Michel Marin / Christian Robert -- 4.1.1. The posterior distribution / Jean-Michel Marin / Christian Robert -- 4.1.2. Bayesian estimates / Jean-Michel Marin / Christian Robert -- 4.1.3. Conjugate prior distributions / Jean-Michel Marin / Christian Robert -- 4.1.4. Noninformative priors / Jean-Michel Marin / Christian Robert -- 4.1.5. Bayesian credible sets / Jean-Michel Marin / Christian Robert -- 4.2. Bayesian discrimination between models / Jean-Michel Marin / Christian Robert -- 4.2.1. The model index as a parameter / Jean-Michel Marin / Christian Robert -- 4.2.2. The Bayes Factor / Jean-Michel Marin / Christian Robert -- 4.2.3. The ban on improper priors / Jean-Michel Marin / Christian Robert -- 4.2.4. The Bayesian Information Criterium / Jean-Michel Marin / Christian Robert -- 4.2.5. Bayesian Model Averaging / Jean-Michel Marin / Christian Robert -- 4.3. The case of linear regression models / Jean-Michel Marin / Christian Robert -- 4.3.1. Conjugate prior / Jean-Michel Marin / Christian Robert -- 4.3.2. Zellner's G prior distribution / Jean-Michel Marin / Christian Robert -- 4.3.3. HPD regions / Jean-Michel Marin / Christian Robert -- 4.3.4. Calculation of evidences and Bayes factors / Jean-Michel Marin / Christian Robert -- 4.3.5. Variable Selection / Jean-Michel Marin / Christian Robert -- 5.1. Some Monte Carlo strategies to approximate the evidence / Jean-Michel Marin / Christian Robert -- 5.1.1. The basic Monte Carlo solution / Jean-Michel Marin / Christian Robert -- 5.1.2. Usual importance sampling approximations / Jean-Michel Marin / Christian Robert -- 5.1.3. The Harmonic mean approximation / Christian Robert / Jean-Michel Marin -- 5.1.4. The Chib's method / Jean-Michel Marin / Christian Robert -- 5.2. The bridge sampling methodology to compare embedded models / Jean-Michel Marin / Christian Robert -- 5.3. A Monte Carlo Markov Chain method for variable selection / Christian Robert / Jean-Michel Marin -- 5.3.1. The Gibbs sampler / Christian Robert / Jean-Michel Marin -- 5.3.2. A Stochastic Search for the Most Likely Model / Christian Robert / Jean-Michel Marin -- 6.1. Motivations / Nicolas Vayatis -- 6.2. Randomness, bless our data! / Nicolas Vayatis -- 6.2.1. A probabilistic view of classification data / Nicolas Vayatis -- 6.2.2. Let the data go: error estimation and model validation / Nicolas Vayatis -- 6.3. Power to the masses: aggregation principles / Nicolas Vayatis -- 6.3.1. Voting and averaging in binary classification / Nicolas Vayatis -- 6.3.2. A lazy way to multi-class classification / Nicolas Vayatis -- 6.3.3. Agreement and averaging in the context of scoring / Nicolas Vayatis -- 6.3.4. From bipartite ranking to K-partite ranking / Nicolas Vayatis -- 6.4. Time for doers: popular aggregation meta-algorithms / Nicolas Vayatis -- 6.4.1. Bagging / Nicolas Vayatis -- 6.4.2. Boosting / Nicolas Vayatis -- 6.4.3. Forests for bipartite ranking and scoring / Nicolas Vayatis -- 6.5. Time for thinkers: Theory of aggregated rules / Nicolas Vayatis -- 6.5.1. Aggregation of classification rules / Nicolas Vayatis -- 6.5.2. Consistency of Forests / Nicolas Vayatis -- 6.5.3. From bipartite consistency to K-partite consistency / Nicolas Vayatis -- 7.1. Mixture models as a many-purpose tool / Christophe Biernacki -- 7.1.1. Starting from applications / Christophe Biernacki -- 7.1.2. The mixture model answer / Christophe Biernacki -- 7.1.3. Classical mixture models / Christophe Biernacki -- 7.1.4. Other models / Christophe Biernacki -- 7.2. Estimation / Christophe Biernacki -- 7.2.1. Overview / Christophe Biernacki -- 7.2.2. Maximum likelihood and variants / Christophe Biernacki -- 7.2.3. Theoretical difficulties related to the likelihood / Christophe Biernacki -- 7.2.4. Estimation algorithms / Christophe Biernacki -- 7.3. Model selection in density estimation / Christophe Biernacki -- 7.3.1. Need to select a model / Christophe Biernacki -- 7.3.2. Frequentist approach and deviance / Christophe Biernacki -- 7.3.3. Bayesian approach and integrated likelihood / Christophe Biernacki -- 7.4. Model selection in (semi-)supervised classification / Christophe Biernacki -- 7.4.1. Need to select a model / Christophe Biernacki -- 7.4.2. Error rates-based criteria / Christophe Biernacki -- 7.4.3. A predictive deviance criterion / Christophe Biernacki -- 7.5. Model selection in clustering / Christophe Biernacki -- 7.5.1. Need to select a model / Christophe Biernacki -- 7.5.2. Partition-based criteria / Christophe Biernacki -- 7.5.3. The Integrated Completed Likelihood criterion / Christophe Biernacki -- 7.6. Experiments on real data sets / Christophe Biernacki -- 7.6.1. BIC: extra-solar planets / Christophe Biernacki -- 7.6.2. AICcond/BIC/AIC/BEC/ecv: benchmark data sets / Christophe Biernacki -- 7.6.3. AICcond/ecvV: textile data set / Christophe Biernacki -- 7.6.4. BIC: social comparison theory / Christophe Biernacki -- 7.6.5. NEC: marketing data / Christophe Biernacki -- 7.6.6. ICL: prostate cancer data / Christophe Biernacki -- 7.6.7. BIC: density estimation in the steel industry / Christophe Biernacki -- 7.6.8. BIC: partitioning communes of Wallonia / Christophe Biernacki -- 7.6.9. ICLbic/BIC: acoustic emission control / Christophe Biernacki -- 7.6.10. ICLbic/ICL/BIC/ILbayes: a seabird data set / Christophe Biernacki -- 7.7. Future methodological challenges / Christophe Biernacki -- 8.1. The concept of minimal penalty / Pascal Massart -- 8.1.1. A small number of models / Pascal Massart -- 8.1.2. A large number of models / Pascal Massart -- 8.2. Data-driven penalties / Pascal Massart -- 8.2.1. From theory to practice / Pascal Massart -- 8.2.2. The slope heuristics / Pascal Massart -- 9.1. Introduction / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.2. HD clustering: Curse or blessing? / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.2.1. HD density estimation: Curse / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.2.2. HD clustering: A mix of curse and blessing / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.2.3. Intermediate conclusion / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.3. Non-canonical models / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.3.1. Gaussian mixture of factor analysers / Christophe Biernacki / Cathy Maugis-Rabusseau -- 9.3.2. HD Gaussian mixture models / Christophe Biernacki / Cathy Maugis-Rabusseau

Owning institution

1. Columbia University Libraries

Note

1. "Société Française de Statistique."
2. "This volume originates from the organization of high-level specialists ... who were all speakers at the 16th bienn[i]al workshop on advanced statistics by the French Statistical Society"--Back cover.

Bibliography (note)

1. Includes bibliographical references and index.