Modeling and Stochastic Learning for Forecasting in High Dimensions

ISBN-13: 9783319187310 / Angielski / Miękka / 2015 / 339 str.

Anestis Antoniadis; Jean-Michel Poggi; Xavier Brossat

Opis

The chapters in this volume stress the need for advances in theoretical understanding to go hand-in-hand with the widespread practical application of forecasting in industry. Forecasting and time series prediction have enjoyed considerable attention over the last few decades, fostered by impressive advances in observational capabilities and measurement procedures. On June 5-7, 2013, an international Workshop on Industry Practices for Forecasting was held in Paris, France, organized and supported by the OSIRIS Department of Electricite de France Research and Development Division. In keeping with tradition, both theoretical statistical results and practical contributions on this active field of statistical research and on forecasting issues in a rapidly evolving industrial environment are presented. The volume reflects the broad spectrum of the conference, including 16 articles contributed by specialists in various areas. The material compiled is broad in scope and ranges from new findings on forecasting in industry and in time series, on nonparametric and functional methods and on on-line machine learning for forecasting, to the latest developments in tools for high dimension and complex data analysis.

Spis treści

1 Short Term Load Forecasting in the Industry for Establishing Consumption Baselines: A French Case.- 2 Confidence intervals and tests for high-dimensional models: a compact review.- 3 Modelling and forecasting daily electricity load via curve linear regression.- 4 Constructing Graphical Models via the Focused Information Criterion.- 5 Nonparametric short term Forecasting electricity consumption with IBR.- 6 Forecasting the electricity consumption by aggregating experts.- 7 Flexible and dynamic modeling of dependencies via copulas.- 8 Operational and online residential baseline estimation.- 9 Forecasting intra day load curves using sparse functional regression.- 10 Modelling and Prediction of Time Series Arising on a Graph.- 11 GAM model based large scale electrical load simulation for smart grids.- 12 Spot volatility estimation for high-frequency data: adaptive estimation in practice.- 13 Time series prediction via aggregation: an oracle bound including numerical cost.- 14 Space-time trajectories of wind power generation: Parametrized precision matrices under a Gaussian copula approach.- 15 Game-theoretically Optimal Reconciliation of Contemporaneous Hierarchical Time Series Forecasts.- 16 The BAGIDIS distance: about a fractal topology, with applications to functional classification and prediction.

O autorze

Anestis Antoniadis is Emeritus Professor at the Department of Applied Mathematics (Laboratoire Jean Kuntzmann), University Joseph Fourier, Grenoble and is also honorary research associate at the Department of Statistical Sciences, University of Cape Town, South Africa. His research interests include wavelet theory, nonparametric function estimation, abstract inference of stochastic processes, statistical pattern recognition, and statistical methodology in meteorology and crystallography. He is a Fellow of the American Statistical Association and of the Institute of Mathematical Statistics and an elected member of the ISI. He has delivered the 2012 Laplace Memorial Lecture in Statistics at the 8th World Congress in Probability and Statistics. Xavier Brossat is a senior research Engineer at Electricity de France in the Department Optimisation, Risks and Statistics for Energy Market within the Research and Development Division. He has participated in several big projects including themes such as Automatic Command of Production Network System and also very short load curve forecasting models. In particular he has participated with several academic and industrial colleagues in developing and adapting methods such as mixtures and aggregation of experts and functional times series prediction to the context of electrical forecasts. He is one of the main organizer of the WIPFOR conference series. Jean-Michel Poggi is Professor of Statistics at University of Paris Descartes and at University Paris-Sud Orsay in France. His main research areas are tree-based methods for classification and regression, nonparametric time series forecasting, wavelet methods and applied statistical modeling in energy and environment fields. His publications combine theoretical and practical contributions together with industrial applications and software development. He is Associate Editor of three journals: Journal of Statistical Software, CSBIGS and Journal de la SFdS. From 2011 to 2013 he was President of the French Statistical Society (SFdS) and, since 2012, he is Vice-President of the Federation of European National Statistical Societies (FENStatS). He is an elected member of the ISI and member of the Board of Directors of the ERS of IASC.

Blurb

The chapters in this volume stress the need for advances in theoretical understanding to go hand-in-hand with the widespread practical application of forecasting in industry.

Forecasting and time series prediction have enjoyed considerable attention over the last few decades, fostered by impressive advances in observational capabilities and measurement procedures. On June 5-7, 2013, an international Workshop on Industry Practices for FORecasting was held in Paris, France, organized and supported by the OSIRIS Department of Electricité de France Research and Development Division.

In keeping with tradition, both theoretical statistical results and practical contributions on this active field of statistical research and on forecasting issues in a rapidly evolving industrial environment are presented. The volume reflects the broad spectrum of the conference, including 16 articles contributed by specialists in various areas. The material compiled is broad in scope and ranges from new findings on forecasting in industry and in time series, on nonparametric and functional methods, and on on-line machine learning for forecasting, to the latest developments in tools for high dimension and complex data analysis.