COMPSTAT 2006 - Proceedings in Computational Statistics - 17th Symposium Held in Rome, Italy, 2006

Preface

Contents

Part I Classification and Clustering

Issues of robustness and high dimensionality in cluster analysis

1 Introduction

2 Multivariate t Distribution

3 ML Estimation of Mixtures of t Components

4 Factor Analysis Model for Dimension Reduction

5 Mixtures of Normal Factor Analyzers

6 Mixtures of t Factor Analyzers

7 Discussion

References

Fuzzy K-medoids clustering models for fuzzy multivariate time trajectories

1 Introduction

2 Fuzzy data time arrays, fuzzy multivariate time trajectories and dissimilarity measures

3 Fuzzy K-means clustering models for fuzzy multivariate time trajectories [ CD03]

4 Fuzzy K-medoids clustering for fuzzy multivariate time trajectories

5 Application

References

Bootstrap methods for measuring classification uncertainty in latent class analysis

1 Introduction

2 Measures of classification uncertainty

3 The bootstrap method

4 Bootstrapping LC models

5 Applications

6 Discussion

References

A robust linear grouping algorithm

1 Introduction

2 Linear Grouping Algorithm

3 Robust Linear Grouping Algorithm

4 Examples

5 Discussion

References

Computing and using the deviance with classification trees

1 Introduction

2 Tree induction principle: an illustrative example

3 Validating the tree descriptive ability

4 Computational aspects

5 Conclusion

References

Estimation procedures for the false discovery rate: a systematic comparison for microarray data

1 Introduction

2 The testing problem

3 The false discovery rate

4 Estimation procedures

5 The data sets

6 Outline of the comparative study

7 Results and conclusions

Acknowledgment

References

A unifying model for biclustering*

1 Illustrative Example

2 Biclustering

3 A Unifying Biclustering Model

4 Data Analysis

5 Concluding Remarks

References

Part II Image Analysis and Signal Processing

Non-rigid image registration using mutual information

1 Introduction

2 Non-rigid registration

3 The mutual information criterion

4 Non-rigid registration using mutual information

5 Validation

References

Musical audio analysis using sparse representations

1 Introduction

2 Finding Sparse Representations

3 Sparse Representations for Music Transcription

4 Source Separation

5 Conclusions

Acknowledgements

References

Robust correspondence recognition for computer vision

1 Introduction

2 Stability and Digraph Kernels

3 Properties of Strict Sub-Kernels

4 A Simple Algorithm for Interval Orientations

5 Discussion

References

Blind superresolution

1 Introduction

2 Mathematical Model

3 Blind Superresolution

4 Experiments

5 Conclusions

Acknowledgment

References

Analysis of Music Time Series

1 Introduction

2 Model building

3 Applied models

4 Studies

5 Conclusion

References

Part III Data Visualization

Tying up the loose ends in simple, multiple, joint correspondence analysis

1 Introduction

2 Basic CA theory

3 Multiple and joint correspondence analysis

4 Data sets used as illustrations

5 Measuring variance and comparing different tables

6 The myth of the influential outlier

7 The scaling problem in CA

8 To rotate or not to rotate

9 Statistical significance of results

10 Loose ends in MCA and JCA

Acknowledgments

References

3 dimensional parallel coordinates plot and its use for variable selection

1 Introduction

2 Parallel coordinates plot and interactive operations

3 3 dimensional parallel coordinates plot

4 Implementation of 3D PCP software

5 Concluding remarks

References

Geospatial distribution of alcohol-related violence in Northern Virginia

1 Introduction

2 Overview of the Model

3 The Data

4 Estimating the Probabilities

5 Geospatial Visualization of Acute Outcomes

6 Conclusions

Acknowledgements

References

Visualization in comparative music research

1 Introduction

2 Music representations

3 Musical databases

4 Musical feature extraction

5 Data mining

6 Examples of visualization of musical collections

7 Conclusion

References

Exploratory modelling analysis: visualizing the value of variables

1 Introduction

2 Example — Florida 2004

3 Selection — More than just Variable Selection

4 Graphics for Variable Selection

5 Small or LARGE Datasets

6 Summary and Outlook

References

Density estimation from streaming data using wavelets

1 Introduction

2 Recursive Formulation

3 Discounting Old Data

4 A Case Study: Internet Header Traffic Data

References

Part IV Multivariate Analysis

Reducing conservatism of exact small-sample methods of inference for discrete data

1 Introduction

2 Small-Sample Inference for Discrete Distributions

3 Ways of Reducing Conservatism

4 Fuzzy Inference Using Discrete Data

5 The Mid-P Quasi-Exact Approach

Acknowledgement

References

Symbolic data analysis: what is it?

1 Symbolic Data

2 Structure

3 Analysis: Symbolic vis-a-vis Classical Approach

4 Conclusion

References

A dimensional reduction method for ordinal three- way contingency table

1 Introduction

2 Decomposing a Non Symmetric Index

3 The Partition of a Predictability Measure

4 Ordinal Three-Way Non Symmetrical Correspondence Analysis

5 Example

References

Operator related to a data matrix: a survey

1 The initial choices

2 Joint analysis of several data matrices (the STATIS method)

3 Principal component analysis with respect to instrumental variables

4 Conclusions

Acknowledgements

References

Factor interval data analysis and its application

1 Introduction

2 Methodology of Interval Data and Its Possible Limitations

3 Methodology of Factor Interval Data and Its Advantages

4 Application in Chinese Stock Markets

5 Conclusion

References

Identifying excessively rounded or truncated data

1 Data

2 DensityModels

3 Asymptotic Behavior

4 Conclusion

Acknowledgements

References

Statistical inference and data mining: false discoveries control

Introduction

1 Data Mining Specificities and Statistical Inference

2 Validation of Interesting Features

3 Controlling UAFWER Using the BS FD Algorithm

4 Experimentation

Conclusion and Perspectives

References

Is ‘Which model . . .?’ the right question?

1 Introduction

2 Preliminaries

3 From choice to synthesis

4 Example

5 Conclusion

References

Use of latent class regression models with a random intercept to remove the effects of the overall response rating level

1 Introduction

2 Description of the cracker case study

3 The LC ordinal regression model with a random intercept

4 Results obtained with the cracker data set

5 General discussion

References

Discrete functional data analysis

1 Introduction

2 Functional Data

3 Difference Operators

4 Detection of Relations among Differences

5 Concluding Remarks

References

Self organizing MAPS: understanding, measuring and reducing variability

1 Introduction

2 Several Approaches Concerning the Preservation of the Topology

3 Understanding Variability of SOM’ Neighbourhood Structure Visualizing Distances between All Classes

4 The R-map Method to Increase SOM Reliability

5 Application: Validating the Number of Units for a SOM Network

6 Conclusion

References

Parameterization and estimation of path models for categorical data

1 Introduction

2 Log-linear, graphical and DAG models

3 DAG models as marginal models

4 Parameterization of DAG models

5 Path models

6 Maximum likelihood estimation

7 An example

References

Latent class model with two latent variables for analysis of count data

1 Introduction

2 Model

3 Analysis of retail market data

References

Part V Web Based Teaching

Challenges concerning web data mining

1 Motivation

2 Challenges Concerning Algorithmic Aspects

3 Conclusions and Further Research

References

e-Learning statistics – a selective review

1 Introduction

2 Modern e-Learning Materials

3 Evaluation

4 Conclusion

References

Quality assurance of web based e-Learning for statistical education

1 Introduction

2 Important Features of the e-StatEdu System

3 Quality Assurance

4 Discussion

Acknowledgement

References

Part VI Algorithms

Genetic algorithms for building double threshold generalized autoregressive conditional heteroscedastic models of time series

1 Introduction

2 The DTGARCHModel

3 A Genetic Algorithm for DTGARCH Model Building

4 Application to Financial Data

5 Conclusions

References

Nonparametric evaluation of matching noise

1 Introduction and preliminaries

2 Statistical framework for matching noise

3 Matching noise for KNN distance hot-deck

4 An important special case: distance hot-deck

5 d0-Kernel hot-deck

6 A comparison among different techniques

References

Subset selection algorithm based on mutual information

1 Introduction

2 Estimation of mutual information using normal mixture

3 Algorithm for subset selection

4 Numerical investigation with real data set

References

Visiting near-optimal solutions using local search algorithms

1 Background and motivation

2 Definitions and notation

3 The ß-acceptable solution probability

4 Visiting a ß-acceptable solution

5 Computational results

6 Conclusions

References

The convergence of optimization based GARCH estimators: theory and application*

1 Introduction

2 Convergence of Optimization Based Estimators

3 Application to GARCH Model

4 Results

5 Conclusions

References

The stochastics of threshold accepting: analysis of an application to the uniform design problem

1 Introduction

2 Formal Framework

3 Results for Uniform Design Implementation

4 Conclusions and Outlook

References

Part VII Robustness

Robust classification with categorical variables

1 Introduction

2 Cluster detection through diagnostic monitoring

3 Performance of the method

4 E-government data

Acknowledgement

References

Multiple group linear discriminant analysis: robustness and error rate

1 Introduction

2 Estimation and Robustness

3 Optimal Error Rate for Three Groups

4 Simulations

5 Conclusions

References

Author Index