BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Cláudia Nunes (CEMAT and IST) DTSTART:20200528T100000Z DTEND:20200528T110000Z DTSTAMP:20260422T212557Z UID:ProbStat/1 DESCRIPTION:Title: Quasi-analytical solution of an investment problem with decreasing invest ment cost due to technological innovations\nby Cláudia Nunes (CEMAT a nd IST) as part of Probability & Statistics (IST-CEMAT\, FC-CEAUL\, ULisb on)\n\n\nAbstract\nIn this talk we address\, in the context of real option s\, an investment problem with two sources of uncertainty: the price (refl ected in the revenue of the firm) and the level of technology. The level o f technology impacts in the investment cost\, that decreases when there is a technology innovation. The price follows a geometric Brownian motion\, whereas the technology innovations are driven by a Poisson process. As a c onsequence\, the investment region may be attained in a continuous way (du e to an increase of the price) or in a discontinuous way (due to a sudden decrease of the investment cost).\n\nFor this optimal stopping problem no analytical solution is known\, and therefore we propose a quasi-analytical method to find an approximated solution that preserves the qualitative fe atures of the exact solution. This method is based on a truncation procedu re and we prove that the truncated solution converges to the solution of t he original problem.\n\nWe provide results for the comparative statics for the investment thresholds. These results show interesting behaviors\, par ticularly\, the investment may be postponed or anticipated with the intens ity of the technology innovations and with their impact on the investment cost.\n\n(joint work with Carlos Oliveira and Rita Pimentel)\n LOCATION:https://researchseminars.org/talk/ProbStat/1/ END:VEVENT BEGIN:VEVENT SUMMARY:Igor Kravchenko (CEMAT and IST) DTSTART:20200618T100000Z DTEND:20200618T110000Z DTSTAMP:20260422T212557Z UID:ProbStat/2 DESCRIPTION:Title: Investment problem with switching modes\nby Igor Kravchenko (CEMAT an d IST) as part of Probability & Statistics (IST-CEMAT\, FC-CEAUL\, ULisbo n)\n\n\nAbstract\nIn this talk we will look at the optimal control problem of a firm that may operate in two different modes\, one being more risky than the other\, in the sense that in case the demand decreases\, the retu rn of the risky mode is lower than with the more conservative mode. On the other side\, in case the demand increases\, the opposite holds. The switc hes between these two alternative modes have associated costs. In both mod es\, there is the option to exit the market.\nWe will focus on two differe nt parameter scenarios\, that describe particular (and somehow extreme) ec onomic situations. In the first scenario\, we assume that the market is ex pected to increase in such a way that once the firm is producing in the mo re risky mode\, it is never optimal to switch to the more conservative one . In the second scenario\, there is a hysteresis region\, where the firm i s waiting in the more risky mode\, in production\, until some drop or incr ease in the demand leads to an exit or changing to the more conservative m ode. This hysteresis region cannot be attained under continuous production .\nWe then address the problem of the optimal time to invest under each si tuation. Depending on the relation between the switching costs (equal or d ifferent from one mode to another)\, it may happen that the firm invests i n the hysteresis region.
\nJoint work with Cláudia Nunes and Carlos O liveira\n LOCATION:https://researchseminars.org/talk/ProbStat/2/ END:VEVENT BEGIN:VEVENT SUMMARY:Maria do Rosário Oliveira (CEMAT and IST) DTSTART:20200625T100000Z DTEND:20200625T110000Z DTSTAMP:20260422T212557Z UID:ProbStat/3 DESCRIPTION:Title: Theoretical foundations of forward feature selection methods based on mu tual information\nby Maria do Rosário Oliveira (CEMAT and IST) as par t of Probability & Statistics (IST-CEMAT\, FC-CEAUL\, ULisbon)\n\n\nAbstr act\nFeature selection problems arise in a variety of applications\, such as microarray analysis\, clinical prediction\, text categorization\, image classification and face recognition\, multi-label learning\, and classifi cation of internet traffic. Among the various classes of methods\, forward feature selection methods based on mutual information have become very po pular and are widely used in practice. However\, comparative evaluations o f these methods have been limited by being based on specific datasets and classifiers. In this talk\, we discuss a theoretical framework that allows evaluating the methods based on their theoretical properties. The estimat ion difficulties of the method’s objective functions will also be addres sed.\n\nThis is a joint work with Francisco Macedo\, António Pacheco\, an d Rui Valadas.\n LOCATION:https://researchseminars.org/talk/ProbStat/3/ END:VEVENT BEGIN:VEVENT SUMMARY:Joaquim Ferreira (Laboratório de Farmacologia Clínica e Terapêu tica\, Faculdade de Medicina\, Universidade de Lisboa) DTSTART:20200723T100000Z DTEND:20200723T110000Z DTSTAMP:20260422T212557Z UID:ProbStat/4 DESCRIPTION:Title: COVID\, uncertainty and clinical trials\nby Joaquim Ferreira (Laborat ório de Farmacologia Clínica e Terapêutica\, Faculdade de Medicina\, Un iversidade de Lisboa) as part of Probability & Statistics (IST-CEMAT\, FC -CEAUL\, ULisbon)\n\n\nAbstract\nThe current COVID-19 pandemic is putting an enormous pressure not just in the society but also in all the scientifi c community.\n\nIf we want to follow a scientific approach to respond to t he doubts and challenges that were generated\, we need to find a balance b etween the most robust data\, the best experimental methodologies to addre ss the new problems and all the uncertainty associated.\n\nIn this present ation we will try to address this balance between best available data\, cl inical research methodology and uncertainty applied to what we know about pandemics\, vaccine development and clinical trials. There will be a parti cular focus on the COVID-19 pandemic data and current research efforts for the development of vaccines and efficacious treatments.\n LOCATION:https://researchseminars.org/talk/ProbStat/4/ END:VEVENT BEGIN:VEVENT SUMMARY:Miguel de Carvalho (University of Edinburgh) DTSTART:20200716T100000Z DTEND:20200716T110000Z DTSTAMP:20260422T212557Z UID:ProbStat/5 DESCRIPTION:Title: Elements of Bayesian geometry\nby Miguel de Carvalho (University of E dinburgh) as part of Probability & Statistics (IST-CEMAT\, FC-CEAUL\, ULi sbon)\n\n\nAbstract\nIn this talk\, I will discuss a geometric interpretat ion to Bayesian inference that will yield a natural measure of the level o f agreement between priors\, likelihoods\, and posteriors. The starting po int for the construction of the proposed geometry is the observation that the marginal likelihood can be regarded as an inner product between the pr ior and the likelihood. A key concept in our geometry is that of compatibi lity\, a measure which is based on the same construction principles as Pea rson correlation\, but which can be used to assess how much the prior agre es with the likelihood\, to gauge the sensitivity of the posterior to the prior\, and to quantify the coherency of the opinions of two experts. Esti mators for all the quantities involved in our geometric setup are discusse d\, which can be directly computed from the posterior simulation output. S ome examples are used to illustrate our methods\, including data related t o on-the-job drug usage\, midge wing length\, and prostate cancer.\n\nJoin t work with G. L. Page and with B. J. Barney\n LOCATION:https://researchseminars.org/talk/ProbStat/5/ END:VEVENT BEGIN:VEVENT SUMMARY:Manuel Cabral Morais (CEMAT and IST) DTSTART:20200514T100000Z DTEND:20200514T110000Z DTSTAMP:20260422T212557Z UID:ProbStat/6 DESCRIPTION:Title: On ARL-unbiased charts to monitor the traffic intensity of a single serve r queue\nby Manuel Cabral Morais (CEMAT and IST) as part of Probabilit y & Statistics (IST-CEMAT\, FC-CEAUL\, ULisbon)\n\n\nAbstract\n

We know too well that the effective operation of a queueing system requires maintaining the traffic intensity at a target value. This important measure of congestion can be monitored b y using control charts\, such as the one found in the seminal work by Bhat and Rao (1972) or more recently in Chen and Zhou (2015). For all intents and purposes\, this paper focus on three control statistics chosen by Mora is and Pacheco (2016) for their simplicity\, recursive and Markovian chara cter:

Since an upward and a dow nward shift in the traffic intensity are associated with a deterioration a nd an improvement (respectively) of the quality of service\, the timely de tection of these changes is an imperative requirement\, hence\, begging fo r the use of ARL-unbiased charts Pignatiello et al. (1995)\, in the sense that they detect any shifts in the traffic intensity sooner than they trig ger a false alarm. In this paper\, we focus on the design of these type of charts for the traffic intensity of the three single server queues mentio ned above.

Joint work with Sven Knoth

\n LOCATION:https://researchseminars.org/talk/ProbStat/6/ END:VEVENT BEGIN:VEVENT SUMMARY:Ismael Lemhadri (Stanford University) DTSTART:20200709T150000Z DTEND:20200709T160000Z DTSTAMP:20260422T212557Z UID:ProbStat/7 DESCRIPTION:Title: LassoNet: A Neural Network with Feature Sparsity\nby Ismael Lemhadri (Stanford University) as part of Probability & Statistics (IST-CEMAT\, FC -CEAUL\, ULisbon)\n\n\nAbstract\nMuch work has been done recently to make neural networks more interpretable\, and one obvious approach is to arrang e for the network to use only a subset of the available features. In linea r models\, Lasso (or $\\ell_1$-regularized) regression assigns zero weight s to the most irrelevant or redundant features\, and is widely used in dat a science. However the Lasso only applies to linear models. Here we introd uce LassoNet\, a neural network framework with global feature selection. O ur approach enforces a hierarchy: specifically a feature can participate i n a hidden unit only if its linear representative is active. Unlike other approaches to feature selection for neural nets\, our method uses a modifi ed objective function with constraints\, and so integrates feature selecti on with the parameter learning directly. As a result\, it delivers an enti re regularization path of solutions with a range of feature sparsity. On s ystematic experiments\, LassoNet significantly outperforms state-of-the-ar t methods for feature selection and regression. The LassoNet method uses p rojected proximal gradient descent\, and generalizes directly to deep netw orks. It can be implemented by adding just a few lines of code to a standa rd neural network.\n LOCATION:https://researchseminars.org/talk/ProbStat/7/ END:VEVENT BEGIN:VEVENT SUMMARY:Conceição Amado (Instituto Superior Técnico and CEMAT) DTSTART:20201001T120000Z DTEND:20201001T130000Z DTSTAMP:20260422T212557Z UID:ProbStat/8 DESCRIPTION:Title: From high dimensional space to a random low dimensional space\nby Con ceição Amado (Instituto Superior Técnico and CEMAT) as part of Probabil ity & Statistics (IST-CEMAT\, FC-CEAUL\, ULisbon)\n\n\nAbstract\nWhat mig ht happen if we have points in a high dimensional space and one decided to project them into a random low dimensional space?\n\nIn this seminar\, we will discuss this subject and will see some simple applications.\n LOCATION:https://researchseminars.org/talk/ProbStat/8/ END:VEVENT BEGIN:VEVENT SUMMARY:Boris Beranger (School of Mathematics and Statistics\, University New South Wales\, Sydney) DTSTART:20201015T100000Z DTEND:20201015T110000Z DTSTAMP:20260422T212557Z UID:ProbStat/10 DESCRIPTION:Title: High-dimensional inference for max-stable processes\nby Boris Berang er (School of Mathematics and Statistics\, University New South Wales\, Sy dney) as part of Probability & Statistics (IST-CEMAT\, FC-CEAUL\, ULisbon )\n\n\nAbstract\nDroughts\, high temperatures and strong winds are key cau ses of the recent bushfires that have touched a major part of the Australi an territory. Such extreme events seem to appear with increasing frequency \, creating an urgent need to better understand the behaviour of extreme e nvironmental phenomena. Max-stable processes are a widely popular tool to model spatial extreme events with several flexible models available in the literature. For inference on max-stable models\, exact likelihood estimat ion becomes quickly computationally intractable as the number of spatial l ocations grows\, limiting their applicability to large study regions or fi ne grids. In this talk\, we introduce two methodologies based on composite likelihoods\, to circumvent this issue. First\, we assume the occurrence times of maxima available in order to incorporate the Stephenson-Tawn conc ept into the composite likelihood framework. Second\, we propose to aggreg ate the information between locations into histograms and to derive a comp osite likelihood variation for these summaries. The significant improvemen ts in performance of each estimation procedures is established through sim ulation studies and illustrated on two temperature datasets from Australia .\n\nJoint seminar CEMAT and CEAUL\n LOCATION:https://researchseminars.org/talk/ProbStat/10/ END:VEVENT BEGIN:VEVENT SUMMARY:Carina Silva (Escola Superior de Tecnologia da Saúde from Lisbon and CEAUL) DTSTART:20201022T120000Z DTEND:20201022T130000Z DTSTAMP:20260422T212557Z UID:ProbStat/11 DESCRIPTION:Title: Impact of OVL Variation on AUC Bias Estimated by Non-parametric Methods< /a>\nby Carina Silva (Escola Superior de Tecnologia da Saúde from Lisbon and CEAUL) as part of Probability & Statistics (IST-CEMAT\, FC-CEAUL\, ULisbon)\n\nAbstract: TBA\n LOCATION:https://researchseminars.org/talk/ProbStat/11/ END:VEVENT BEGIN:VEVENT SUMMARY:Jorge Milhazes de Freitas (Faculty of Sciences of the University of Porto and CMUP) DTSTART:20201028T130000Z DTEND:20201028T140000Z DTSTAMP:20260422T212557Z UID:ProbStat/12 DESCRIPTION:Title: Enriched functional limit theorems for dynamical systems\nby Jorge M ilhazes de Freitas (Faculty of Sciences of the University of Porto and CM UP) as part of Probability & Statistics (IST-CEMAT\, FC-CEAUL\, ULisbon)\ n\nAbstract: TBA\n LOCATION:https://researchseminars.org/talk/ProbStat/12/ END:VEVENT BEGIN:VEVENT SUMMARY:Ana Paula Martins (Faculty of Sciences of University of Beira Inte rior and CMA) DTSTART:20201112T130000Z DTEND:20201112T140000Z DTSTAMP:20260422T212557Z UID:ProbStat/13 DESCRIPTION:Title: Extremes modelling of imputed missing data under a periodic control\ nby Ana Paula Martins (Faculty of Sciences of University of Beira Interior and CMA) as part of Probability & Statistics (IST-CEMAT\, FC-CEAUL\, ULi sbon)\n\nAbstract: TBA\n LOCATION:https://researchseminars.org/talk/ProbStat/13/ END:VEVENT END:VCALENDAR