- Random - Mailing list — Dipartimento di Matematica, UniPI

UNIBO Statistics Seminars: Davide Ferrari, Free University of Bolzano/Bozen
by Simone Giannerini 24 Oct '22

24 Oct '22

We are glad to announce the following Statistics Seminar: Friday, October 28, 11.30 a.m. Davide Ferrari (Free University of Bozen-Bolzano) Fast Construction of Optimal Composite Likelihoods Abstract A composite likelihood is a combination of low-dimensional likelihood objects useful in applications where the data have complex structure. Although composite likelihood construction is a crucial aspect influencing both computing and statistical properties of the resulting estimator, currently there does not seem to exist a universal rule to combine low-dimensional likelihood objects that is statistically justified and fast in execution. This paper develops a methodology to select and combine the most informative low-dimensional likelihoods from a large set of candidates while carrying out parameter estimation. The new procedure minimizes the distance between composite likelihood and full likelihood scores subject to a constraint representing the afforded computing cost. The selected composite likelihood is sparse in the sense that it contains a relatively small number of informative sub-likelihoods while the noisy terms are dropped. The resulting estimator is found to have asymptotic variance close to that of the minimum-variance estimator constructed using all the low-dimensional likelihoods. Preprint on Statistica Sinica https://www3.stat.sinica.edu.tw/ss_newpaper/SS-2021-0235_na.pdf Room 2, Piazza Scaravilli 2, Bologna Link for attending the event: Fai clic qui per partecipare alla riunione<https://teams.microsoft.com/l/meetup-join/19%3ameeting_ZjY4MjE4MDQtYmM3Mi00…> Contact person: Simone Giannerini _________________________________________________________________ PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY A Theme issue ‘DNA as information<https://royalsocietypublishing.org/toc/rsta/374/2063>’ edited by Julyan H.E. Cartwright, Simone Giannerini and Diego L. González _________________________________________________________________ Simone Giannerini Dipartimento di Scienze Statistiche "Paolo Fortunati" Universita' di Bologna Via delle belle arti 41 - 40126 Bologna, ITALY Tel: +39 051 2098262 Fax: +39 051 232153 https://www.unibo.it/sitoweb/simone.giannerini/ ______________________________________________________ ________________________________

1 0

LUISS Math Seminars: Peter Bank, Thu 27 October
by Sara 21 Oct '22

21 Oct '22

Dear all, I am pleased to announce the upcoming seminar by Peter Bank, TU Berlin: https://economiaefinanza.luiss.it/seminar/2022/09/23/what-if-we-knew-what-f… The seminar will take place in Room 106 at 12pm, at the address below. You can also attend the seminar online, at the link: *Room 106:* https://luiss.webex.com/luiss/j.php?MTID=mc76ccac202efab1b18aea6a847fc1457 *User:* w_guest(a)luiss.it *Password: *ilLGhdI522Q Very best, Sara -- Sara Biagini, Professor of Mathematical Finance Department of Economics and Finance LUISS Guido Carli https://www.luiss.it/ Address: Viale Romania, 32 - 00197 Roma Web: http://sites.google.com/site/sarabiagini/

1 0

I: Statistique mathématique: Post-Doc à Orsay
by Rita Giuliano 20 Oct '22

20 Oct '22

Inoltro per eventuali interessati Rita Giuliano ________________________________ ________________________________ De: "Yohann" <yohann.de-castro(a)ec-lyon.fr> À: icj(a)math.univ-lyon1.fr Envoyé: Mercredi 19 Octobre 2022 10:37:28 Objet: [icj] Post-Doc à Orsay Bonjour, J’ai le plaisir de vous faire part d'une annonce de post-doc à Orsay en statistique mathématique. Cordialement Yohann De Castro _______________________________________________ ICJ mailing list ICJ(a)math.univ-lyon1.fr https://es.sonicurlprotection-fra.com/click?PV=2&MSGID=20221019085709100024…

1 0

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