Séminaire Probabilités et Statistique

Having access to one’s own money is facilitated by banking channels, namely; branches, ATMs, call centres, online and mobile banking as well as the post office. Whilst banks provide channels, individual experiences of access are not homogenous. Indeed, socio-economic and geographical considerations lead to vastly varied access. With digital banking having permeated society, individuals across the UK use physical network points of access – such as branches or ATMs – less than previously. As such, the cost of running such network points have become relatively more expensive and banks have started retreating their physical access point presence. Whilst this has been anecdotally reported in mainstream media, there has been no quantitative analysis on the impact of said closures.

As such, we propose a simple methodology to identify critical points of access. Our hope is that regulators and industry players can then use this framework to ascertain integral nodes the the UK’s banking channel network. The framework considers all channels available to an individual, in respect to their place of residence. The distance to the closest point of access is calculated as well as the impact of the closure of that access point. The impact is measured as the difference in distance between the closest and second-closest point of access, reflecting an increase in the difficulty of banking access

Exploratory spatial data analysis at both UK and regional level showed strong spatial patterns of the points of access to banking ; significant rural/urban clusters could be identified as well as a North/South divide which we need to explore further. No significant association was found between distance metrics and income and employment. Despite data limitations, the indicators used in this study can be used to identify areas vulnerable to the closure of the last points of access. We learned that the majority of the infrastructure for access is no longer operated by banks. In this context, it becomes even more critical to maintain and monitor a dynamic map of access and therefore we recommend more transparency on location, capability and capacity of the points of access from all players, as well as on broadband availability and quality from telecom providers. Retail banking access should be treated as a joined-up system so that territorial coverage can be ensured, such that entire communities are not accidentally excluded from participation in the economy.

Je vais vous parler d’un travail effectué en collaboration avec Pierre Hodara and Eva Löcherbach. On s’intéresse à  un processus de Hawkes à  mémoire variable. On considère un modèle de neurones en interaction o๠le potentiel des membrane des neurones est décrit comme un Processus de Markov déterministe par morceaux (notés PDMP) à  valeurs dans $mathbb{R}^N, $ o๠$ N$ décrit le nombre de neurones. Un drift déterministe attire chaque potentiel de la membrane du neurone vers un potentiel à  l’équilibre $m$. Lorsqu’un neurone saute, le potentiel de sa membrane est réinitialisé à  zéro et les autres gagnent $frac{1}{N}.$ On s’intéresse à  l’estimation du taux de sauts d’un neurone basée sur l’observation du potentiel des $N$ neurones observés jusqu’à  un temps $t$. On va étudier un estimateur à  noyaux Nadaraya-Watson pour le taux de sauts et on va établir la vitesse de convergence dans $L^2 .$

In this talk we shall review a few well known problems related to the distribution of the values of the Riemann zeta function on the critical line and see how one can use probabilistic models and ideas to hopefully gain better insights into such problems. The main focus of the talk will be on the GUE conjecture and the convergence of some random analytic functions (coming from random matrix theory).

Nous étudions un graphe aléatoire motivé par des questions de biologie
évolutive. Ce graphe aléatoire est défini comme la distribution
stationnaire d’une chaîne de Markov soumise aux deux transitions
suivantes : la duplication de sommet, lors de laquelle, à  taux constant,
un sommet est choisi uniformément, déconnecté de tous ses voisins puis
reconnecté à  un autre sommet choisi uniformément ainsi qu’aux voisins de
ce second sommet; et la perte d’arêtes, qui consiste à  faire disparaître
les arêtes du graphe à  taux constant par arête.

Une approche coalescente permet d’obtenir des formules explicites pour
les premiers moments de plusieurs invariants de graphe tels que le
nombre d’arêtes ou le nombre de sous-graphes complets d’ordre k. Nous
donnerons également une formule explicite pour la distribution des
degrés et des bornes asymptotiques sur le nombre de composantes connexes.

I am going to talk about a continuum model of directed polymer in random environment. The law of the polymer is defined as the Brownian motion conditioned to survive among space-time Poissonian disasters. This model is well-studied in the positive temperature regime. However, at zero-temperature, even the existence of the free energy has not been proved. Stefan Junk and I have proved that the free energy exists and is continuous at zero-temperature.

Le but de cet exposé est de présenter une introduction à  un thème de
théorie ergodique : l’équivalence orbitale stable. Ce problème issu de
la classification des algèbres de von Neumann, se propose d’étudier les
systèmes dynamiques modulo une relation plus faible que la conjugaison:
l’équivalence orbitale. Deux tels systèmes sont orbitalement équivalent
s’il y a un isomorphisme préservant les orbites de chacun des systèmes.
Nous verrons comment les principaux invariants ergodiques (entropie,
mesure invariante, spectre,…) varient au sein d’une même classe
d’équivalence orbitale.

Ce congrès annuel de la Société Française de Statistique, qui va durer toute la seamine,

Spatial modeling in Astronomy can benefit from visualization techniques and spatial manipulation using Geographic Information Systems (G.I.S.) techniques. Preliminary analysis of stellar population characteristics in M83 galaxy using the emitted radiation source in different wavelengths shows a spatial aggregation model of stellar populations and clusters that are hierarchically organized in proximity of galaxy’s spiral arms. A review on stellar population and cluster characteristics (Krumholz et al., 2018) shows that their definition and organization characteristics vary on orders of several magnitudes, therefore the need to analyze clustering models using several mathematical models as well as G.I.S. clusters tools are foreseen here (a set of weighted features, identifies statistically significant hot spots, cold spots, and spatial outliers using the Anselin Local Moran’s I statistic). We aim to check if the clustering is related to a theoretical model (the density wave theory), where local to regional organization of stellar populations that belong to main sequence can form the hierarchically organized structures along the M83 galaxy’s spiral arms.