Accueil > Actualités > Evénements > Workshop Data science and forecast validation , Aussois 9-12 March 2020

Workshop Data science and forecast validation , Aussois 9-12 March 2020

Date : 09-03-2020 au 12-03-2020

Today, there exists a plethora of numerical models that can produce ensemble forecasts or projections.
Are they reliable with respect to observations? capable of reproducing complex spatial or temporal features? should they be combined and how?

In this context, the field of statistical science has produced recent advances in  forecast verification methods and has highlighted some roadblocks and research questions.


To address some of these issues,  we are pleased to announce the workshop "Assessment of ensemble forecasts” (VALPRED 2) that will take place in Aussois on 9-12 March 2020.

This data science event is sponsored by the IPSL-Climate Graduate School* and by the Mathematics Jacques Hadamard Foundation. It is also one of the planned 2020 actions of the IPSL-SAMA group  (Statistiques pour l'Analyse, la Modélisation et l'Assimilation).  

This  four-day workshop will deal  with the topic of validation of ensemble forecasting.
In a nutshell, our goal is to mix around 30 researchers, statisticians, meteorologists, climatologists, energy providers and other end-users  who have a common interest in probability forecasting, especially in the assessment of different forecasts.
During this workshop,  our state of knowledge on ensemble forecast validation will be reviewed. The probability scoring tools for such assessments  will be discussed.

As stochastic models and methods are key to representing and controlling  uncertainties,  they  will also receive particular attention in this context.


The non-exhaustive list of potential topics will be:
- scoring rules theory,
- forecasting methods,
- extreme value theory,
- decision theory.


The two main applications will be weather/climate forecast and energy consumption forecasts, but other applications are welcome.

As the workshop starting date is approaching quickly, let me know if you are interested by this event asap.
If so, please send me a tentative title, short abstract and a CV before the 5th of February.  

If selected, fundings can cover lodging and food expenses at the CNRS-CAES Aussois center from Monday  to Thursday lunch.


Priorities will be given to young IPSL researchers interested in statistics/data science/machine learning  techniques.


Philippe Naveau (philippe.naveau@lsce.ipsl.fr) and Olivier Wintenberger  



PS : The website http://wintenberger.fr/VALPRED.html will be also updated as more info will come in.
* This work benefited from the French state aid managed by the ANR under the "Investissements d'avenir" programme with the reference ANR-11-IDEX-0004 - 17-EURE-0006

Retour à la liste evénements