This book presents the latest research on the statistical analysis of functional, high-dimensional and other complex data, addressing methodological and computational aspects, as well as real-world applications. It covers topics like classification, confidence bands, density estimation, depth, diagnostic tests, dimension reduction, estimation on manifolds, high- and infinite-dimensional statistics, inference on functional data, networks, operatorial statistics, prediction, regression, robustness, sequential learning, small-ball probability, smoothing, spatial data, testing, and topological object data analysis, and includes applications in automobile engineering, criminology, drawing recognition, economics, environmetrics, medicine, mobile phone data, spectrometrics and urban environments. The book gathers selected, refereed contributions presented at the Fifth International Workshop on Functional and Operatorial Statistics (IWFOS) in Brno, Czech Republic. The workshop was originally to be held on June 24-26, 2020, but had to be postponed as a consequence of the COVID-19 pandemic. Initiated by the Working Group on Functional and Operatorial Statistics at the University of Toulouse in 2008, the IWFOS workshops provide a forum to discuss the latest trends and advances in functional statistics and related fields, and foster the exchange of ideas and international collaboration in the field.
Les mer
Preface.- List of Contributors .- 1 An introduction to the (postponed) 5th edition of the International Workshop on Functional and Operatorial Statistics.- 2 Analysis of Telecom Italia Mobile Phone Data by Space-time Regression with Differential Regularization.- 3 Some Numerical Test on the Convergence Rates of Regression with Differential Regularization.- 4 Learning with Signatures.- 5 About the Complexity Function in Small-ball Probability Factorization.- 6 Principal Components Analysis of a Cyclostationary Random Function.- 7 Level Set and Density Estimation on Manifolds.- 8 Pseudo-metrics as Interesting Tool in Nonparametric Functional Regression.- 9 Testing a Specification Form in Single Functional Index Model.- 10 A New Method for Ordering Functional Data and its Application to Diagnostic Test.- 11 A Functional Data Analysis Approach to the Estimation of Densities over Complex Regions.- 12 A Conformal Approach for Distribution-free Prediction of Functional Data.- 13 G-Lasso  Network Analysis for Functional Data.- 14 Modelling Functional Data with High-dimensional Error Structure.- 15 Goodness-of-fit Tests for Functional Linear Models Based on Integrated Projections.- 16 From High-dimensional to Functional Data: Stringing Via Manifold Learning.- 17 Functional Two-sample Tests Based on Empirical Characteristic Functionals.- 18 Some Remarks on the Nelson–Siegel Model.- 19 Modeling the Effect of Recurrent Events on Time-to-event Processes by Means of Functional Data.- 20 On Robust Training of Regression Neural Networks.- 21 Simultaneous Inference for Function-valued Parameters: a Fast and Fair Approach.- 22 Single Functional Index Model under Responses MAR and Dependent Observations.- 23 O2S2 for the Geodata Deluge .- 24 Riemannian Distances between Covariance Operators and Gaussian Processes.- 25 Depth in Infinite-dimensional Spaces.- 26 Variable Selection in Semiparametric Bi-functional Models.- 27 Local Inference for Functional Data Controlling the Functional False Discovery Rate.- 28 Optimum Scale Selection for 3D Point Cloud Classification through Distance Correlation.- 29 Generalized Functional Partially Linear Single-index Models.- 30 Functional Outlier Detection through Probabilistic Modelling.- 31 Topological Object Data Analysis Methods with an Application to Medical Imaging .- 32 Distribution-free Pointwise Adjusted %-values for Functional Hypotheses.- Authors Index.
Les mer
This book presents the latest research on the statistical analysis of functional, high-dimensional and other complex data, addressing methodological and computational aspects, as well as real-world applications. It covers topics like classification, confidence bands, density estimation, depth, diagnostic tests, dimension reduction, estimation on manifolds, high- and infinite-dimensional statistics, inference on functional data, networks, operatorial statistics, prediction, regression, robustness, sequential learning, small-ball probability, smoothing, spatial data, testing, and topological object data analysis, and includes applications in automobile engineering, criminology, drawing recognition, economics, environmetrics, medicine, mobile phone data, spectrometrics and urban environments. The book gathers selected, refereed contributions presented at the Fifth International Workshop on Functional and Operatorial Statistics (IWFOS) in Brno, Czech Republic. The workshop was originally to be held on June 24-26, 2020, but had to be postponed as a consequence of the COVID-19 pandemic. Initiated by the Working Group on Functional and Operatorial Statistics at the University of Toulouse in 2008, the IWFOS workshops provide a forum to discuss the latest trends and advances in functional statistics and related fields, and foster the exchange of ideas and international collaboration in the field.
Les mer
Presents the latest advances in functional and high-dimensional statistics Covers methodological and computational aspects as well as applications Appeals to a wide audience, from theoretical and computationally oriented statisticians to experimental scientists
Les mer

Produktdetaljer

ISBN
9783030477554
Publisert
2020-06-20
Utgiver
Vendor
Springer Nature Switzerland AG
Høyde
235 mm
Bredde
155 mm
Aldersnivå
Research, P, 06
Språk
Product language
Engelsk
Format
Product format
Innbundet

Biographical note

Germán Aneiros is an Associate Professor of Statistics at the University of A Coruña, Spain. His research focuses on statistical inference for functional data, including sparse semi-parametric regression models, selection of impact points in a curve, bootstrap procedures and functional prediction of electricity demand and price. He is an Associate Editor of the journal Computational Statistics.

Ivana Horová is a Full Professor of Applied Mathematics at Masaryk University, Brno, Czech Republic. Her research focuses on nonparametric statistical methods, particularly multivariate kernel smoothing and its applications. She is a co-author of a monograph on kernel smoothing in MATLAB. She was a Guest Editor of the special issue Computational Environmetrics in the journal Environmetrics in 2009.

Marie Hušková is a Full Professor of Mathematical Statistics at Charles University, Prague, Czech Republic. She is the author of more than 130 scientific papers, mainly on asymptotic statistics, nonparametric and multivariate statistics and change-point problems. She is an Associate Editor of the journals Metrika, Statistics, and Sequential Analysis, and is a former Associate Editor of the Journal of Statistical Planning and Inference and REVSTAT. She is an elected member of ISI and a fellow of IMS. For several years, she was the chair of the European Regional Committee of the Bernoulli Society and a member of the Council of ISI.

Philippe Vieu is a Full professor at Paul Sabatier University, Toulouse, France.  He is well known for his numerous achievements in fields such as nonparametric statistics and functional statistics. He was an editor of previous IWFOS proceedings and several special issues on functional and nonparametric statistics. Currently, he is a Co-Editor of the journal Computational Statistics, and an Associate Editor of the Journal of Nonparametric Statistics, TEST, Statistics & Probability Letters and the Journal of Multivariate Analysis.