Random

random@lists.dm.unipi.it

47 participants
5251 discussions

Seminar: Catherine Matias 26 October: “Properties of the stochastic approximation EM algorithm”
by Fulvia Pennoni 20 Oct '22

20 Oct '22

Dear all, We are glad to invite you to the *seminar* that will take place on the *26th of October, at 14.30*, in presence in Room *Aula Seminari DiSMeQ 4026, building U7* Civitas, 4th floor, *University of Milano-Bicocca*, Via Bicocca degli Arcimboldi 8, 20126 Milano. *Catherine Matias* <http://cmatias.perso.math.cnrs.fr/>from *Centre National de la Recherche Scientifique (CNRS), Sorbonne University, Paris* will present a seminar on “*Properties of the stochastic approximation EM algorithm with the mini-batch sampling*” (see abstract below). The seminar is also available online at the following link: https://unimib.webex.com/unimib-it/j.php?MTID=m414dda85e475f906edae49b2a8ee… <https://unimib.webex.com/unimib-it/j.php?MTID=m414dda85e475f906edae49b2a8ee…> Passcode: NmEQKpfG535 (66375734 from phones) You are invited to forward the event to your students, PhDs and colleagues who may be interested in the Seminar. Kind Regards, Fulvia Pennoni On behalf of the Department of Statistics and Quantitative Methods /--------------/ *Speaker*: Catherine Matias from Centre National de la Recherche Scientifique (CNRS), Sorbonne University, Paris *Title*: /Properties of the stochastic approximation EM algorithm with the mini-batch sampling/ *Abstract*: To deal with very large datasets a mini-batch version of the Monte Carlo Markov Chain Stochastic Approximation Expectation–Maximization algorithm for general latent variable models is proposed. For exponential models the algo rithm is shown to be convergent under classical conditions as the number of iterations increases. Numerical experiments illustrate the performance of the mini-batch algorithm in various models. In particular, we highlight that mini-batch sampling results in an important speed-up of the convergence of the sequence of estimators generated by the algorithm. Moreover, insights on the effect of the mini-batch size on the limit distribution are presented. Finally, we illustrate how to use mini-batch sampling in practice to improve results when a constraint on the computing time is given. -- Fulvia Pennoni Ph.D. Dpt of Statistics and Quantitative Methods University of Milano-Bicocca

1 0

Open call - Research Fellow at University of Oslo
by Giulia Di Nunno 20 Oct '22

20 Oct '22

Dear All, Let me bring to your attention the following position as Research Fellow (postdoc 18 months) at the Department of Mathematics, University of Oslo: https://www.jobbnorge.no/en/available-jobs/job/233446/research-fellow-at-th… Deadline for application: October 30, 2022 The Research fellow is expect to work within the scope of STORM - Stochastics for Time-Space Risk Models. In particular the research work will focus on the development of the theory and the application of models non-necessarily Markovian suitable for the modelling of market date and associated risks. More details are given in the call. Please feel free to distribute this ad further. Best regards, Giulia Di Nunno Professor Department of Mathematics University of Oslo, Norway

1 0

Seminar - Raffaele Argiento - 20 Oct, 2:30 pm - Florence
by Monia Lupparelli 19 Oct '22

19 Oct '22

Dear Colleagues, the Department of Statistics, Computer Science, Application of University of Florence is glad to invite you to the following seminar: Raffaele Argiento (Università degli Studi Bergamo) Thursday 20 October 2022, 2:30 pm Dipartimento di Statistica, Informatica, Applicazioni "G. Parenti”, room 205 Viale Morgagni, 59, Firenze Hierarchical normalized finite point process: predictive structure and clustering Abstract Almost surely discrete random probability measures have received close attention in the Bayesian nonparametric community. They have been used to model populations of individuals or latent parameters (in the mixture model setting) composed of unfixed species with unknown proportions. In this framework, data are usually assumed to be exchangeable. However, the latter assumption is not appropriate when data are divided in multiple groups which may share the same species. If so, partially exchangeability accommodates the dependence across populations. Best regards, Monia Lupparelli ----------------------------------------- Monia Lupparelli, PhD Department of Statistics, Computer Science, Applications University of Florence Viale Morgagni 59, 50134 Florence, IT Phone +39 055 2751579

1 0

Seminar: Erik Lindström 24 October: Feature Selection in Jump Models
by Fulvia Pennoni 19 Oct '22

19 Oct '22

Dear all, We are glad to invite you to the *seminar* that will take place on the *24th of October, at 14.30*, in presence in Room *Aula Seminari DiSMeQ 4026, building U7* Civitas, 4th floor, *University of Milano-Bicocca*, Via Bicocca degli Arcimboldi 8, 20126 Milano. *Erik Lindström <https://www.maths.lu.se/staff/erik-lindstroem/>*from *Lund University* will present a seminar on “*Feature Selection in Jump Models*” (see abstract below). The seminar is also available online at the following link: https://unimib.webex.com/unimib/j.php?MTID=m7d32c71febb43ee95aabde1ca5428861 You are invited to forward the event to your students, PhDs and colleagues who may be interested in the Seminar. Kind Regards, Fulvia Pennoni On behalf of the Department of Statistics and Quantitative Methods /--------------/ *Speaker*: Erik Lindström from Lund University *Title*: /Feature Selection in Jump Models/ *Abstract*: Hidden Markov models are a popular choice for inferring the hidden state of financial markets as the states often are interpreted as market regimes. This interpretation implicitly assumes that the underlying state sequence has a certain level of persistence. However, when a hidden Markov model is misspecified or misestimated, it often leads to unrealistically rapid switching dynamics. We propose a novel estimation approach based on clustering temporal features while penalizing jumps. The advantages of the proposed jump estimator include that it learns the hidden state sequence and model parameters simultaneously while providing control over the transition rate, it is less sensitive to initialization, it performs better when the number of states increases, and it is robust to misspecified conditional distributions. In addition, feature selection is necessary in high-dimensional settings where the number of features is large compared to the number of observations and the underlying states differ only with respect to a subset of the features. We develop and implement a coordinate descent algorithm that alternates between selecting the features and estimating the model parameters and state sequence, which scales to large data sets with large numbers of (noisy) features. We demonstrate the usefulness of the proposed framework by comparing it with a number of other methods on both simulated and real data in the form of financial returns, protein sequences, and text. The resulting sparse jump model outperforms all other methods considered and is remarkably robust to noise. -- Fulvia Pennoni Ph.D. Dpt of Statistics and Quantitative Methods University of Milano-Bicocca

1 0

Mathematical Methods for Climate: Link to Minicourse by D. Faranda
by Francesco Grotto 18 Oct '22

18 Oct '22

Good afternoon, you can find below the link for the second part of the minicourse by D. Faranda; the lecture is slightly delayed due to technical issues. Best FG https://cnrs.zoom.us/j/92979982747?pwd=UVF1QkJqOXdjWVBrc2V6M2tMck54QT09

1 0

Seminar by René Aid (Paris Dauphine) in Padova (hybrid mode)
by Giorgia Callegaro 18 Oct '22

18 Oct '22

Dear all, you're invited to the next seminar in Probability and Finance, that will take place on Friday 28th October, at 11.30, in hybrid mode at the Mathematics Department of the University of Padova. More details: * *Speaker*: *René A**ïd (Paris Dauphine)* * *Date and time*: 28th October 2022, 11.30 * *Room (Torre Archimede)*: 2BC30 * *Zoom link*: please find it at https://www.math.unipd.it/~bianchi/seminari/ * *Title*: *A model of optimal incentives for renewable investments* * *Abstract*: We analyze the optimal regulatory incentives to foster the development of non-emissive electricity generation when the demand for power is served either by a monopoly or by two competing firms. The regulator wishes to encourage renewable investments to limit carbon emissions, while simultaneously reducing intermittency of the total energy production. We provide the expression of the optimal contract for each market structure. We show how the market structure affects the complexity of the optimal incentive mechanism. A joint work with Annika Kemper and Nizar Touzi. See you soon, thanks, Giorgia Callegaro -- Giorgia Callegaro Associate Professor Department of Mathematics - University of Padova Via Trieste 63 , I-35121 Padova - ITALY Tel: +39-0498271481 Fax: +39-0498271499 E-Mail: gcallega(a)math.unipd.it <https://webmail.math.unipd.it/horde3/imp/message.php?mailbox=Sent&index=598#> Personal web-page: https://sites.google.com/site/giogiocallegaro/Home

1 0

Seminar Marinucci @Bocconi - 21st Oct at 12:00
by Giacomo Zanella 18 Oct '22

18 Oct '22

We are glad to announce the following hybrid seminar: Thursday, 21st October 2022 12:00, Room AS01 + Zoom Bocconi University, Via Roentgen 1, floor -1 Speaker: Domenico Marinucci (Università di Roma Tor Vergata) Title: Asymptotics for Spherical Functional Autoregressions Abstract: In this talk, we investigate a class of spherical functional autoregressive processes, and we discuss the estimation of the corresponding autoregressive kernels. In particular, we first establish a consistency result (in sup and mean-square norm), then a quantitative central limit theorem (in Wasserstein distance), and finally a weak convergence result, under more restrictive regularity conditions. We shall also hint at possible extensions, in particular nonparametric testing for stationarity, isotropy and Gaussianity. Based on joint works with Alessia Caponera and Anna Vidotto. Zoom information: link: https://unibocconi-it.zoom.us/j/92683143026?pwd=Q3plSDVPRzAvcjhQUllBemNPVjV… Meeting ID: 926 8314 3026 Passcode: 123456 Best regards, Giacomo Zanella

1 1

Mathematical Methods for Climate Science: Minicourse and Seminar by D. Faranda in Pisa
by Francesco Grotto 17 Oct '22

17 Oct '22

Dear colleagues, in the context of the project "Metodi matematici per la scienza del clima" at Dipartimento di Matematica in Pisa, D. Faranda (LSCE laboratory, Paris Saclay) will be giving: -a minicourse, "Understanding, researching and telling the story of climate extreme events" on 17-18th November at Dipartimento di Matematica, - a seminar, "Atmospheric circulation compounds anthropogenic warming and its impacts in Europe", on 19th November at Scuola Normale Superiore. All the information can be found at the following link: https://clima.dm.unipi.it/2022/10/12/minicourse-and-seminar-by-d-faranda-ls… Best Francesco Grotto

1 2

Seminar in Statistics - MARIE KRATZ
by Pierpaolo De Blasi 14 Oct '22

14 Oct '22

SEMINARS IN STATISTICS @ COLLEGIO CARLO ALBERTO <https://www.carloalberto.org/events/category/seminars/seminars-in-statistic…> Venerdi 21 Ottobre 2022, alle ore 12.00, presso il Collegio Carlo Alberto, in Piazza Arbarello 8, Torino, si terrà il seguente seminario: ------------------------------------------------ Speaker: *Marie Kratz* (ESSEC Business School, CREAR risk research center) Title: *Pro-Cyclicality Beyond Business Cycles: The Case of Traditional Risk Measurements* Abstract: There is an accepted idea that risk measurements are pro-cyclical: in times of crisis, they overestimate the future risk, while they underestimate it in quiet times. We lay down a simple and efficient methodology to evaluate the amount of pro-cyclicality in the way financial institutions measure risk. To do so, we introduce a new indicator based on the Sample Quantile Process (a dynamic extension of Value-at-Risk), conditioned on realized volatility. Using this framework, we prove that pro-cyclicality is inherent in risk measure estimates based on historical data. We identify two main factors explaining this pro-cyclical behavior: the clustering and return-to-the-mean of volatility, as it could have been anticipated but not yet quantified, and, more surprisingly, the very way risk is measured, even in a world with constant volatility, though the empirical magnitude of the mean-reversion is greater than what would be observed in that special case. We develop CLTs and FCLT’s for functionals of quantile and dispersion estimators to support theoretically those empirical findings. This is a joint work with Dr. Marcel Bräutigam and Dr. Michel Dacorogna. ------------------------------------------------ Sarà possibile seguire il seminario anche in streaming: Join Zoom Meeting <https://us02web.zoom.us/j/85751895656?pwd=dTBiQTAzd3BoODRoZmQzMlZWTDZIQT09> Meeting ID: 857 5189 5656 Passcode: 166674 Il seminario è organizzato dalla "de Castro" Statistics Initiative www.carloalberto.org/stats -- Pierpaolo De Blasi University of Torino & Collegio Carlo Alberto carloalberto.org/pdeblasi <https://sites.google.com/a/carloalberto.org/pdeblasi/>

1 0

Roxana Dumitrescu's talk, Milan 19/10/22
by Luciano Campi 14 Oct '22

14 Oct '22

Dear all I am happy to announce the following seminar by Roxana Dumitrescu (King’s College London), you’ll find details on date and venue below the abstract Title: Control and stopping mean-field games: the linear programming approach Abstract: In this talk, we present recent results on the linear programming approach to mean-field games in a general setting. This relaxed control approach allows to prove existence results under weak assumptions, and lends itself well to numerical implementation. We consider mean-field game problems where the representative agent chooses both the optimal control and the optimal time to exit the game, where the instantaneous reward function and the coefficients of the state process may depend on the distribution of the other agents. Furthermore, we establish the equivalence between mean-field games equilibria obtained by the linear programming approach and the ones obtained via other approaches used in the previous literature. We then present a fictious play algorithm to approximate the mean-field game population dynamics in the context of the linear programming approach. Finally, we give an application of the theoretical and numerical contributions introduced in the first part of the talk to an entry-exit game in electricity markets. The talk is based on several works, joint with R. Aïd, G. Bouveret, M. Leutscher and P. Tankov. Date and time: 19 October 2022, 11am Venue: Sala di Rappresentanza, Departimento di Matematica, Università degli Studi di Milano, Via Saldini 50, 20133 Milano. You are all welcome to attend All the best Luciano -------------------------------------------------------- Luciano Campi Full Professor University of Milan Department of Mathematics Email: luciano.campi(a)unimi.it Web: sites.google.com/site/lucianocampi1

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

Random