Martedi'
10 gennaio 2017 |
ore 11:00 | Daniele Cappelletti U. Wisconsin - Madison, USA |
Reaction network in the deterministic and the stochastic settings Mathematical models for chemical reaction networks are widely used in biochemistry, as well as in other fields. The aim of the models is to predict the dynamics of a collection of reactants that undergo chemical transformations. There exist two standard modeling regimes: a deterministic and a stochastic one. These regimes are chosen case by case in accordance to what is believed to be more appropriate. It is natural to wonder whether the dynamics of the two different models are linked, and whether properties of one model can shed light on the behavior of the other one. Some connections between the two modelling regimes have been known for forty years, and new ones have been pointed out recently. However, many open questions remain, as under what circumstances the boundedness of the trajectories of the deterministic model implies positive recurrence for the stochastic model. |
Martedi' 17 Gennaio 2017 |
ore 16:00 | Lidia Sacchetto Universita' degli Studi di Torino |
Using penalized regression models to select a microRNAs signature for prostate cancer |
Martedi' 17 Gennaio 2017 |
ore 16:45 | Luca Grassano Politecnico di Torino & Roche France |
Correction methods for unmeasured confounding in non interventional studies |
Lunedi' 23 gennaio 2017 |
ore 11:00 | Gabor Lente U. Debrecen, Ungheria |
Stochastic effects and the autocatalytic chemical systems The usual mathematical approach of chemical kinetics relies on concentrations, which are assumed to be continuous functions of time. However, it is now well understood that when very small amounts of substance are involved, the particulate nature of matter makes this approach untenable and the computationally often more demanding stochastic kinetics must be used as an alternative. The talk will argue that the use of stochastic kinetics may be inevitable in interpreting observed phenomena in autocatalytic systems even if the amounts of substance involved are quite macroscopic. Such experimentally confirmed stochastic observations in chemistry include seemingly random distributions of enantiomers in the Soai reaction, and large fluctuations in the clock time of certain Landolt-type reactions, and extinction in certain enzymatic reactions. These experiments were made in systems in which strong autocatalysis had been confirmed earlier. Autocatalysis is a kinetic phenomenon that involves positive feedback. When the feedback is very strong compared to other processes, the kinetic role of just a handful of product molecules may be significant enough to influence the overall reaction, which provides exactly the conditions where the use of stochastic kinetics is inevitable. This was demonstrated earlier by the successful use of the stochastic approach to interpret enantiomeric excess and Landolt time distributions. These earlier results are now developed into a unified theory for autocatalytic processes. It is also important to decide when the use of conventional kinetics is acceptable for autocatalytic reactions. Stochastic maps will be shown to be a very useful aid in deciding this question. |
Giovedi' 09 febbraio 2017 |
ore 12:00 | Volker Kraft JMP Sr. Academic Ambassador |
JMP Demo: Design and Analysis of Experiments (Physical and Simulated) |
Martedi' 14 marzo 2017 |
ore 16:00 | Elvira Erhardt Politecnico and Universita' di Torino |
OPTIMISING THE DRUG RELEASE OF A TRANSDERMAL PATCH The goal is the investigation of the influence of a transdermal patch drug release to optimise its exposure. To achieve this research aim, several steps have to be conducted. Modelling the in-vitro release (considered as a surrogate for the input function) as Weibull absorption and the three-compartment model with first-order elimination (fitted to the IV PK data), were the first steps undertaken. A combination of both models serves as the in-vitro in-vivo model for the controlled release data and the fit of the resulting models is compared to the individual data.. The parameters of the compartmental model and the input function were estimated by non-linear mixed models in R and MONOLIX, and are currently being extended towards a Bayesian framework in STAN. |
Martedi' 14 marzo 2017 |
ore 17:00 | Lidia Sacchetto Politecnico e Universita' di Torino |
CART: Classification and Regression Trees In this seminar I will present the classical approach to tree-based decision methods, following what done by Breiman and colleagues and focusing on classification trees. I will introduce the basics of pruning and I will present an example using the software R, on data provided by the Cancer Genomic Lab of the Tempia Foundation in Biella. I will also provide some general information on bagging, random forests, and boosting procedures, as well as on how CART handle missing values. |
Giovedi' 06 aprile 2017 |
ore 15:00 | Fortunato Pesarin Department of Statistical Sciences, University of Padova |
Nonparametric Union-Intersection Approach in Multivariate Permutation Tests |
Martedi' 11 Aprile 2017 |
ore 17:15 | Mauro Gasparini Politecnico di Torino |
The basics of quantitative research methods. This is seminar geared towards master-level students in the applied sciences approaching for the first time Quantitative Research (as when writing a master's thesis). The students will have a minimal quantitative background but not necessarily formal statistical education. I will provide a non-technical overview of modern methods for data analysis using Statistics and some Machine Learning. The seminar will be in English. |
Giovedi' 20 aprile 2017 |
ore 15:00 | Ulrike Grömping Beuth University of Applied Sciences, Berlin |
Understanding the confounding structure of factorial experiments |
Giovedi' 04 maggio 2017 |
ore 15:00 | Giovanni Pistone de Castro Statistics - Collegio Carlo Alberto, Moncalieri (TO) |
Wasserstein geometry of the Gaussian model |
Giovedi' 11 maggio 2017 |
ore 16:00 | Andrea Romeo Intesa San Paolo, Torino |
Multivariate factor-based processes with Sato margins |
Venerdi' 06 ottobre 2017 |
ore 15:00 | Alan Agresti Distinguished Professor Emeritus, University of Florida |
Some Issues in Generalized Linear Modeling This talk discusses several topics pertaining to generalized linear modeling. With focus on categorical data, the topics include (1) bias in using ordinary linear models with ordinal categorical response data, (2) interpreting effects with nonlinear link functions, (3) cautions in using Wald inference (tests and confidence intervals) when effects are large or near the boundary of the parameter space, and (4) the behavior and choice of residuals for GLMs. I will present few new research results, but these topics got my attention while I was writing the book "Foundations of Linear and Generalized Linear Models," recently published by Wiley. |
Back to Probability, Statistics and Optimization group
This page is maintained by Roberto Fontana;
Mauro Gasparini;
and Barbara Trivellato;
Last modified: 31/03/17