VISITING SCHOLARS OF THE PARIS SEINE INITIATIVE
- Luc BAUWENS
Research project - Cryptocurrencies have attracted public attention in the last months due to the erratic behavior in the valuation of Bitcoin, in particular. The research will focus on the econometric modeling of the returns of the most important cryptocurrencies and the relation between these time series and the corresponding series of returns of traditional financial and commodity markets. The research will enable us to know to what extentcryptocurrencies differ from traditional financial instruments and whether or not they offer additional diversification and hedging opportunities to investors. The econometric modeling will be based on multivariate volatility models known as dynamic conditional correlation models.
- Seminar : February 8, 2018
Research Seminar in Econometrics and Statistics, Department of Information Systems, Decision Sciences and Statistics & ESSEC Research Center:
“The Factorial Hidden Markov Volatility Model”
- Conference : March 29, 2018
- Conference : April 11, 2018
“Recent Developments in Time Series Econometrics”
website: to be announced later
- Seminar : February 8, 2018
- Indranil BISWAS
Indranil Biswas is a professor at the School of Mathematics of Tata Institute of Fundamental Research at Mumbai. His topics of research include Algebraic Geometry,
Analytic Geometry, Topology and Mathematical Physics. He has more than five hundred publications in mathematical journals. He has collaborated with more than 160 mathematicians.
Research project -
The following works were done during the period of visit:
1. Automorphism group of principal bundles, Levi reduction and invariant connections.
This is jointly written with Francois-Xavier Machu (Cergy-Pontoise).
2. Principal co-Higgs bundles on the projective line. This is jointed written with
Oscar Garcia-Prada (ICMAT, Madrid), Jacques Hurtubise (McGill University)
and Steven Rayan (University of Saskatchewan).
3. Branched holomorphic Cartan geometry on Sasakian manifolds. This is jointly
written with Sorin Dumitrescu (Nice) and Georg Schumacher (Marburg).
(Two other works should be ready by the end of the week.)
- Sveva CORRADO
The Laboratory she runs is recognized as one of the most active in Italy for research and training in geohistory of sedimentary basins, but application of thermal maturity of organic matter to other geological issue have been developed. The main research topics developed in the Lab concern
- Thermal and tectonic evolution of fold-and-thrust belts
- Geohistory and thermal history of sedimentary basins for Petroleum system assessment
- Cap rock and reservoir rock quality assessment for geothermal exploration
- Pyroclastic flow temperatures assessment for volcanic risk mitigation.
Research project - Research areas are widespread in various geodynamic settings, active and fossils, all over the world (Italy – Main Cenozoic foreland basins and Mesozoic passive margins deformed in the Apennines and Sicily fold-and-thrust belt; Antartica – Ross Sea; Brazil – Sao Francisco Basin; Namibia – offshore passive margin; Poland – Lublin Basin, Baltic Basin; Spain – Paleozoic Aragon-Bearn Basin, Hecho Basin; Western Pyrenees; Iran – Zagros Mts. (Fars, Dezful Embayement), Mts Alborz; Ukraine – Carpathian fold-and-thrust belt and Podolie foreland; Morocco – Rif orogen; Italy – Northern Latium geothermal area; Argentina – Rosario de La Frontera geothermal area; Portugal (Azores: Fogo, Volcano); Italy - Ercolanum: Vesuvius Volcano; Indonesia - Merapi Volcano; Mexico - Colima Volcano.
In the last five years, Sveva Corrado’s team has been developing a new strategy for the assessment of thermal maturity of organic matter dispersed in sedimentary successions for applications spanning from energy industry to natural risk assessment, integrating also Raman and FTIR spectroscopy.
- Veaceslav COROPCEANU
Research project - Organic solar cells are attracting significant interest due to a number of valuable features such as low cost, low-environmental impact, ﬂexibility and large-area manufacturing capability. Although, organic solar cells currently reached power conversion efficiencies up to 13%, these values remain substantially lower than those in silicon or perovskite solar cells. In order to develop new organic photovoltaic materials with improved efficiency, an in-depth understanding of the fundamental mechanisms that define device performance must be reached. A fundamental issue under much debate in the organic photovoltaic literature relates to the role of excitonic and charge-transfer triplet states. In this project we aim to develop a computational approach capable of providing reliable splitting energies and intrinsic lifetimes of the singlet and triplet charge-transfer states in organic photovoltaic materials. We will also investigate the impact of excitonic and charge-transfer triplet states on the geminate and non-geminate recombination processes.
Impact of static and dynamic disorder on electronic processes in organic materials
- Jorge BERNAL DEL NOZAL
Research project - One of the main limitations of the development and validation of intelligent systems for health is the lack of public annotated datasets in which to test the performance of automatic methods. To cope with this, several efforts have been undertaken to organize international challenges in which researchers share their knowledge and assess the performance of their methodologies over a common public validation framework. As part of my collaboration with ETIS lab at ENSEA, we have organized two international challenges at the main conference of our research domain (MICCAI) on the topic of Gastrointestinal Image ANALYSIS. During the research stage at ENSEA, I will collaborate on the analysis of the results extracted from the challenge held in September 2017 at Quebec, Canada. This analysis will result in the preparation of two different journal articles summarizing the main findings discovered from the challenge; these two articles will be written and coordinated jointly between Jorge Bernal (UAB-CVC) and Aymeric Histace (ETIS lab, ASTRE team, CNRS, ENSEA). Apart from this, I will also continue my collaboration with the team led by Aymeric Histace on the development and validation of real time methods for automatic polyp detection in colonoscopy videos, aiming to progress towards an effective deployment of our technology in the exploration room.
- Conference : Smart Videocolonoscopy, February 13th, 2018 @ 17h30, ENSEA
- Konstantinos FOKIANOS
He was a visiting Assistant Professor of Statistics at The Ohio State University, USA, for period of 2.5 years and has international collaborations with many institutions all over Europe and USA. He was invited by Ludwig Maximilians University, National Institutes of Health, EPFL, University Cergy-Pontoise and University of Bergen and more recently by TU Dortmund for extended visits. He has also several shorter visits to other top academic institutions.
His research interests are focused on the analysis and methodology for time series data and in semiparametric models. His recent focus is on the analysis of integer valued time series. He is co-author, with B. Kedem, of the book Regression Models for Time Series Analysis published by Wiley, 2002. He has co-edited two volumes and he is the author of around 60 peer-reviewed articles. He is an elected member of the International Statistical Institute since 2005 and a series Editor for the Springer collection Frontiers in Probability and the Statistical Sciences. Additionally he is in the editorial boards of Statistical Modelling, Journal of Time Series Analysis and Statistics.
Research project - Count time series refer to data observed over regular time intervals and take integer values. For example, consider the daily number of patients admitted to a hospital or the number of transactions of some stocks, per minute. These simple examples show that measurements might fluctuate according to different observational times. Hence, we aim on introducing mixture distributions for modelling such phenomena. During the lifetime of this project we will be studying the development of new statistical models for regression analysis of mixture count time series data. Furthermore, we will be investigating the issue of model selection for count time series. Mixture models imply different behavior of the observed process at different time regimes. These problems are challenging and difficult to be addressed and there is no satisfactory answer, to the best of our knowledge. For instance, simple fitting and statistical analysis of such models can be quite complicated. We plan to contribute by introducing new models, study their dependence properties and develop statistical inference. Furthermore, while for standard time series the literature of the model selection techniques is large and quite comprehensive, little work has been done in the context of count time series. We envision that we will address, at least partially, this important problem. Overall this work contributes further to the development of time series methods in terms of theory and applications.
- Advanced Courses:
Multivariate Time Series: This class addresses the problems of modeling and inference for multivariate time series. There will be several examplesfrom diverse fields, like medicine, biology, finance and other.Multivariate time series analysis provides several tools and methods for analyzing data observed in multiple measurements having temporal and cross-sectional dependence. The goal is to identify a better understanding of the dynamic relationship between variablesand improve accuracy of prediction.
We will use real data examples and the R language to discuss the following topics:
1. Basic concepts of multivariate time series
2. Stationary vector autoregressive processes
3. Vector autoregressive moving-average time series
This class will meet 7 times for three hours each time
We will organize jointly with P. Doukhan the conference on Non-stationarity (see the following URL).
- Miron KAUFMAN
Research project - Our “social physics” project applies statistical physics techniques to multi-group social conflict. It will expand to several groups the results we obtained for two groups. Individuals in each group have an attitude ranging between collaborative, very open to negotiating an agreement to inclined to protracted conflict due to extreme adherence to the group’s position. We quantify the noise as a “social temperature” T. We assume everyone interacts with everyone in time within their own group and across groups, as on an Erdös-Renyi network. The Hamiltonian H of the interactions depends on the attitude variables. We use the Boltzmann probability weight, exp(-H/T), to compute the probability distributions for attitudes. We explore by means of Monte Carlo simulations effects of the network topology on the qualitative behavior of the model. Its predictions include temporal oscillations of the attitudes towards negotiation or conflict. Monte Carlo simulations exhibit chaotic time dependence of the mean attitudes. We illustrated the model’s use with the 2016 US presidential elections and the Brexit vote. Before the outcome has materialized, the model can help a group devise or alter its strategy in response to the dynamics at work, by generating possible scenarios. Either group could ask what-if questions that can assist in selecting and altering in time a strategy that will be wise for a range of scenarios instead of just one predicted possibility. This anticipatory approach is conducive to robust decisions that can withstand more contextual challenges than decisions based on predicted futures.
- Sanda KAUFMAN
Research project - Our “social physics” project applies statistical physics techniques to multi-group social conflict. It will expand to several groups the results we obtained for two groups. Individuals in each group have an attitude ranging between collaborative, very open to negotiating an agreement to incline to protracted conflict due to extreme adherence to the group’s position. We quantify the noise as a “social temperature” T. We assume everyone interacts with everyone in time within their own group and across groups, as on an Erdös-Renyi network. The Hamiltonian H of the interactions depends on the attitude variables. We use the Boltzmann probability weight, exp(-H/T), to compute the probability distributions for attitudes. We explore by means of Monte Carlo simulations effects of the network topology on the qualitative behavior of the model. Its predictions include temporal oscillations of the attitudes towards negotiation or conflict. Monte Carlo simulations exhibit chaotic time dependence of the mean attitudes. We illustrated the model’s use with the 2016 US presidential elections and the Brexit vote. Before the outcome has materialized, the model can help a group devise or alter its strategy in response to the dynamics at work, by generating possible scenarios. Either group could ask what-if questions that can assist in selecting and altering in time a strategy that will be wise for a range of scenarios instead of just one predicted possibility. This anticipatory approach is conducive to robust decisions that can withstand more contextual challenges than decisions based on predicted futures.
Seminar : Presentation of current research on anticipating the spatial distribution of businesses in a regional space along multiple years, using a dynamic network model.
Conference : Participation atNetSci, Paris
- Joshua SKEWES
Research project - While at the IAS, Dr Skewes will work closely with Maciej Workiewicz, Assistant Professor of Management at ESSEC business school, to develop theoretical models of how organizations make decisions, learn, and adapt to changing environments. The focus of the project will be on how different forms of group organization and structure can enable or impede a group’s ability to adapt to a changing environment. The specific question of the project will be how, with the same set of individuals, can we best structure a group to respond to and capitalize on change in the world? We will approach this question from a biological basis, and develop new models of group and organizational design inspired by knowledge of neural network structure and neural plasticity from the brain and cognitive sciences.