le 13 février 2019
Publié le 26 mai 2021 Mis à jour le 21 février 2022

Non stationary days



Those non-stationary days were made possible thanks to labex MME-DII, ESSEC, CREST, IAS, TELECOM, SAMM, , SFDS and to the various organisers, Pierre Alquier, Jean Marc-Bardet, Paul Doukhan, Olga Klopp, Nicolas Marie, Jean-Luc Prigent, François Roueff, and Jean Michel Zakoian.

The techniques of time series are widely used for the prediction of random phenomena. Other related questions such as for example calibration, resampling or risk management are other important issues.

Stationarity is usually used as a hypothesis because ergodicity is the basis of any asymptotic theory for large samples. Our goal is to omit this condition to describe non-stationary models and methods.

To allow a regular and deep work we preferred to organize four 1-day international conferences dealing with various aspects related to this problem better than a single conference. These days are intended to paint a complete picture of the methods implemented for this purpose. To this aim, we invited the main world specialists of the discipline.

December 12, 2018 - Amphi Jade, Télécom Paristech

February 13, 2019 - ESSEC la Défense

March 7, 2019 - Amphi Hermite, IHP

June 3, 2019 - Université Paris Panthéon Sorbonne:  MSE, floor 6

Read more

Contact: Paul Doukhan