Date
3 months : November 2021 to March 2022.
Short CV:
Prof. Valdério Anselmo Reisen is graduated in mathematics from the Federal University of Espírito Santo (Brazil), Master’s degree in Statistics from the University of Campinas (Brazil), and a Ph.D. in Mathematics – Department of Mathematics, University of Manchester Institute of Science and Technology (UK) and visitor researcher in many institutions in Brazil and abroad. He works in several research groups in Brazil and abroad in time series problems in different areas of application.
Laboratoire
L2S, CentraleSupélec – Université Paris-Saclay
Research interests
Time series analysis (in the time and frequency domains), forecasting, econometric modelling, bootstrap, robustness in time series, unit root processes, counting processes, long-memory, cointegration, periodic processes multivariate time series, environmental and econometrics data analysis, linear and non-linear regression with time series covariates, quantile regression, mixed models, multivariate analysis for temporal correlated data, factorial analysis, hight dimension methods, wavelets.
Main research contributions relevant to DigiCosme
Statistical learning refers to a set of tools for modeling and understanding complex datasets. The statistical methods, in the case of univariate and multivariate data, are crucial in most fundamental steps of this feld. The impact of statistics in data science is discussed recently by Weihs et al.(2018). Time series are very common in studies of observational data. The connection between time series methods and multivariate statistics techniques becomes a very powerful tool to deal with in data science, especially when this also involves robust methods. In this context, the courses proposed below have the aims at addressing the use of time series and regression models in multivariate techniques with robust methodologies in the steps in data science and/or statistical learning.
These courses will be taught at the Master level in CentraleSupélec – Université Paris-Saclay and in AgroParisTech.
Proposed short-courses:
1. Univariate techniques in time series : time domain and frequency approaches
2. Robust methods in multivariate and univariate time series
The first course will be held online on 31st January, from 10-12 am, here :
The second course will be held online on 25th February, from 10-12 am, here:
Proposed seminars (the dates will be available soon):
1. Robust dimension methods in time series and applications.
2. Robust M-estimator in time series and applications
3. Periodic INAR processes and applications
Course:
An introduction to Time series models; A course for the first-year master students.
December 3, from 1:30 PM to 4:45 PM (including 15mn break), 3h course
December 10, from 1:30 PM to 3:15 PM for a second course
Participation in mid-term exams in CentraleSupélec – Université Paris-Saclay
Supervision of PhD thesis from the L2S laboratory, CentraleSupélec – Université Paris-Saclay.