Section: New Results

Curves classification, denoising and forecasting

Participants : Jairo Cugliari, Émilie Devijver, Pascal Massart, Jean-Michel Poggi, Vincent Thouvenot.

In collaboration with Farouk Mhamdi and Meriem Jaidane (ENIT, Tunis, Tunisia), Jean-Michel Poggi proposed a method for trend extraction from seasonal time series through the Empirical Mode Decomposition (EMD). Experimental comparison of trend extraction based on EMD, X11, X12 and Hodrick Prescott filter are conducted. First results show the eligibility of the blind EMD trend extraction method. Tunisian real peak load is also used to illustrate the extraction of the intrinsic trend.

Jean-Michel Poggi was the supervisor (with A. Antoniadis) of the PhD Thesis of Jairo Cugliari-Duhalde which takes place in a CIFRE convention with EDF. It was strongly related to the use of wavelets together with curves clustering in order to perform accurate load consumption forecasting. The thesis contains methodological and applied aspects linked to the electrical context as well as theoretical ones by introducing external variables in the context of nonparametric forecasting time series. See http://hal.archives-ouvertes.fr/docs/00/78/82/49/PDF/cugliari-jma.pdf and http://hal.inria.fr/docs/00/55/99/39/PDF/RR-7515.pdf The industrial post-doc of Jairo Cugliari, funded by EDF, explores three aspects of this model that complement the original methodology: first, the construction of a confidence interval for the predictor function, second, the flexibility and simplicity of the model to provide, without extra effort, forecasts horizons further and further away and finally, and third: study of the ability to provide good predictions in the presence of subtle signal nonstationarities induced by loss of customers coming from various scenarios, see http://hal.archives-ouvertes.fr/docs/00/81/49/24/PDF/kwf-suite.pdf

Jean-Michel Poggi, co-supervising with Anestis Antoniadis (Université Joseph Fourier Grenoble) the PhD thesis of Vincent Thouvenot, funded by a CIFRE with EDF. The industrial motivation of this work is the recent development of new technologies for measuring power consumption by EDF to acquire consumption data for different mesh network. The thesis will focus on the development of new statistical methods for predicting power consumption by exploiting the different levels of aggregation of network data collection. From the mathematical point of view, the work is to develop generalized additive models for this type of kind of aggregated data for the modeling of functional data, associating closely nonparametric estimation and variable selection using various penalization methods.

Jean-Michel Poggi and Pascal Massart are the co-advisors of the PhD thesis of Émilie Devijver, strongly motivated by the same kind of industrial forecasting problems in electricity, is dedicated to curves clustering for the prediction. A natural framework to explore this question is mixture of regression models for functional data. The theoretical subject of the thesis is to extend to functional data the recent work by Bühlmann et al. dealing with the simultaneous estimation of mixture regression models in the scalar case using Lasso type methods. Of course, it will be based on the technical tools of the work of Caroline Meynet (which completes her thesis Orsay under the direction of P. Massart), which deals with the clustering of functional data using Lasso methods choosing simultaneously number of clusters and selecting significant wavelet coefficients.