Yannig Goude, EDF
Vendredi 16 mars 2012
ENGREF,
Adaptive GAM models for Day-Ahead and Intra-Day Electricity Consumption Forecasts
Matthieu Cornec, Fanny Mikol
Insee, CREST
DREES
ENSAE, Amphi 1
Jeudi 9 Février 2012
"Nowcasting GDP directional change with an application to French business survey data".
Emile RICHARD
1000mercis (en CIFRE) et au CMLA-ENS Cachan
Vendredi 20 janvier 2012
Prediction in dynamic graph sequences
Jairo Cugliari
Postdoc at SELECT team, INRIA
Travail realise en these à EDF
Lieu ENGREF
12h30-13-30
Amphi 7rdc
16 déc
Titre
Prévision non paramétrique de processus à valeurs fonctionnelles.
amphi b 208 2eme etage
Sylvain Robbiano
Télécom Paristech
4 novembre
Titre
" ranking binaire et agrégation pour le cas multi-classes"
Lieu ENGREF
12h30-13-30
Amphi 7rdc
16 déc
Groupe de travail sur la prévision entre universitaires et professionnels "non universitaires" (entreprises, institutions publiques).
lundi 13 février 2012
Présentation vendredi 16 mars 2012 : Adaptive GAM models for Day-Ahead and Intra-Day Electricity Consumption Forecasts
Lieu ENGREF
Salle 7
12h30-13h30
Vendredi 16 mars 2012
Yannig Goude
EDF R&D
Adaptive GAM models for Day-Ahead and Intra-Day Electricity Consumption Forecasts
Generalized Additive Models have been investigated recently to forecasts day-ahead electricity consumptions at EDF R&D. These models achieve an interesting trade-off between accuracy of forecasts and adaptation to different data sets thanks to their semi-parametric structures. We propose here a new method based on QR decomposition (joined work with S. Wood) to learn this models on-line as we receive new data. This allows GAM models to react to smooth changes in the data generation process: economic crisis, loss or gain of customers… We illustrate it on different data sets and real forecasts.
Salle 7
12h30-13h30
Vendredi 16 mars 2012
Yannig Goude
EDF R&D
Adaptive GAM models for Day-Ahead and Intra-Day Electricity Consumption Forecasts
Generalized Additive Models have been investigated recently to forecasts day-ahead electricity consumptions at EDF R&D. These models achieve an interesting trade-off between accuracy of forecasts and adaptation to different data sets thanks to their semi-parametric structures. We propose here a new method based on QR decomposition (joined work with S. Wood) to learn this models on-line as we receive new data. This allows GAM models to react to smooth changes in the data generation process: economic crisis, loss or gain of customers… We illustrate it on different data sets and real forecasts.
Adaptive gam
View more presentations from cornec
Inscription à :
Articles (Atom)