Loading...
18 results
Search Results
Now showing 1 - 10 of 18
- Program Book: 2023 IMS International Conference on Statistics and Data Science (ICSDS)Publication . Gomes, Maria Ivette; Oliveira, Teresa A.; Oliveira, Amilcar; Pestana, Pedro Duarte; Xu, MinIn response to the call from the 2021 IMS (Institute of Mathematical Statistics) Survey report to expand membership from emerging areas of data science, underrepresented groups, and from regions outside of North America, the IMS has launched the annual IMS International Conference on Statistics and Data Science (ICSDS). Following the success of 2022 ICSDS in Florence Italy, December 13-16. 2022, the second and this 2023 ICSDS is held on December 18-21, in Lisbon, Portugal. In addition to plenary sessions, invited, contributed and poster sessions, ICSDS offers a student paper competition for 12 Student Travel Awards. There are also Junior Researcher Support Funds for travel support for junior researchers. We gratefully acknowledge here the generous support for both awards from the funds of Industry Friends of IMS (IFoIMS). Students and young researchers are strongly encouraged to participate and utilize this support. The ICSDS conference has been thoughtfully organized to provide platforms to facilitate discovery dissemination and foster collaborations among researchers from a wide range of research and practice areas in statistics and data science, and from academia, industry and government. We are gratified to be able to welcome to the ICSDS more than 550 participants coming from more than 40 countries. The goal of the ICSDS is to afford such a diverse crowd a stimulating setting for exchanging ideas on the developments of modern statistics and data science, broadly defined, in all aspects of theory, methods and applications. The Local Organizing Committee in Lisbon, the beautiful capital city of Portugal, has worked hard to find the modern Cultural Center of Belém (CCB) in Belém area for us to host the ICSDS this year. CCB is located near where Tagus River meets the Atlantic Ocean, and it is surrounded by many magnificent landmarks, including Tower of Belém and Mosteiro dos Jerónimos. This location conveniently affords the conference participants rich social programs, including conference tours to Mosteiro dos Jerónimos and Tower of Belém, both are UNESCO World Heritage Sites, and are historical and architectural treasures for the world. We are also able to organize a conference banquet in the charming Casa do Alentejo to enjoy the local cuisine and the renowned Portuguese traditional Fado music performance. Participants are also highly encouraged to take advantage of the close proximity to “Pastéis de Belém” from CCB to enjoy the world famous Portuguese tarts. Many other wonderful things to do, eat and see in Lisbon can also be found in https://www.timeout.com/lisbon. We wish that the 2023 ICSDS proves to be another productive conference that will successfully cultivate more fruitful exchanges and collaborations for years to come. Acknowledgement: Needless to say, an international conference of this scale, with its coverage of wide-ranging subjects and size of broad participants from various disciplines across the world, would not have been possible without the collective efforts of many. We would like to thank the program committee, of 50 members from 29 countries, for helping establish the rich program. The Local Organizing Committee: Eunice Carrasquinha (Co-chair, CEAUL-FCUL), Ivette Gomes (Co-chair, CEAUL-FCUL), Tiago Marques (University of St Andrews, UK), Teresa A. Oliveira (Co-chair, CEAUL and Universidade Aberta), Soraia Pereira (CEAUL, Universidade de Lisboa), Giovani Silva (IST, CEAUL, Universidade de Lisboa), Lisete Sousa (Universidade de Lisboa), have contributed their tremendous effort and time in every aspect of this endeavor, from helping organize sessions, finding conference venue to all the logistics up to the very last minute, including banquet and tours. In particular, Ivette and Teresa brilliantly turned the messy and long conference program into a beautifully organized program book. Our heartfelt thanks also go to Elyse Gustafson (IMS Executive Director) for her working overtime on the financial issues and other related formalities on behalf of the IMS, and to Arlene Gray (Administrator, ICSDS) for her patiently taking care of the nonstop inquiries and tracking numerous responses and requests from the participants and the conference organizing team. Finally, we would be remiss not to acknowledge the invaluable contributions behind the scenes from Min Xu (IMS, Rutgers University), from setting up and managing the conference website, negotiating IT support with the conference venue, to setting up the program and readying all the slides. He bravely and efficiently met head on all kinds of unexpected challenges, technical as well as personal. It suffices to say that Min did all the heavy-lifting to make the conference program a reality for us all to enjoy.
- Provas de Agregação em Matemática: lição, relatório e curriculumPublication . Oliveira, Amilcar
- An overview of the systemic risk measuresPublication . Basílio, Jorge; Oliveira, Amilcar; Mahmoudvand, RahimSystemic risk is a specific type of risk that refers to the risk of an complex system to be affect or even collapse due to individual action taken by the agents that compounds that complex system. The goals of this work is based on an axiomatic approach establish a critic description of the most relevant methods used in the determination of systemic risk and identify advantages and disadvantages associated to those methods.
- Approximating the distribution of the product of two normally distributed random variablesPublication . Seijas-Macias, J. Antonio; Oliveira, Amilcar; Oliveira, Teresa A.; Leiva, VictorThe distribution of the product of two normally distributed random variables has been an open problem from the early years in the XXth century. First approaches tried to determinate the mathematical and statistical properties of the distribution of such a product using different types of functions. Recently, an improvement in computational techniques has performed new approaches for calculating related integrals by using numerical integration. Another approach is to adopt any other distribution to approximate the probability density function of this product. The skew-normal distribution is a generalization of the normal distribution which considers skewness making it flexible. In this work, we approximate the distribution of the product of two normally distributed random variables using a type of skew-normal distribution. The influence of the parameters of the two normal distributions on the approximation is explored. When one of the normally distributed variables has an inverse coefficient of variation greater than one, our approximation performs better than when both normally distributed variables have inverse coefficients of variation less than one. A graphical analysis visually shows the superiority of our approach in relation to other approaches proposed in the literature on the topic.
- Modeling water level fluctuation in river basins using singular spectrum analysisPublication . Oliveira, Amilcar; Sarmento, CarlaWater scarcity affected 29% of the EU territory during at least one season in 2019. In the face of climate change, it is very important to understand the risk of water scarcity. Water scarcity is becoming a growing problem in southern European countries, such as Portugal. In 2019, Portugal, faced one of the most significant water scarcity conditions in the EU-27 on the seasonal scale (seasonal WEI 66%).The main objective of this work is to study the water level fluctuation in river basins, in order to predict the risks of lack of water. The study area is located in 29 reservoirs from different river basins in Portugal. The collected data refer to the period from November 1993 to August 2022, with a total number of records of 9686. We started by improving the quality of the data and built a monthly time series of the volume of water stored. Next, we analyzed the time series using Singular Spectrum Analysis (SSA), which is a nonparametric technique for analyzing time series.
- Fitting heavy Tail distributions with mixture modelsPublication . Basílio, Jorge; Oliveira, AmilcarThe normal probability distribution as assumption for financial returns have been recognized as inappropriate, and a source of inaccurate estimation of Value at Risk. Empirical evidence also have been shown that financial returns shows a more accentuated leptokurtic distribution when compared with a Normal distribution and also skewed. This is usually a cause of underestimated values of VaR, specially when the quantiles are very low. Therefore it is necessary to focus on the tail of the distribution and identify models to fit that behavior. We will highlight the differences between the quality of fitting in the tails of the distribution and the fitting for all the distribution. This work compares and interprets the results obtained by applying mixture models as a method to estimate the behavior on the extremes for heavy tail data distributions. This results will be then used to describe an analytical solution of VaR under mixture models.
- Big data sets in environmental studiesPublication . Oliveira, AmilcarBig Data datasets for environmental studies play a crucial role in understanding, monitoring and addressing large-scale environmental issues. Big Data datasets for environmental studies deal with huge volumes of data coming from various sources such as satellites, remote sensors, weather stations, sensor networks and mobile devices. These datasets include detailed information on climate change, biodiversity, air quality, water resources and other environmental parameters. Integrating and analyzing data from different sources allows for a more comprehensive understanding of environmental standards and helps in making informed decisions. The generation of environmental data occurs in real time, especially with the increased use of sensors and continuous monitoring technologies. The ability to handle the velocity of data is essential for detecting rapid changes in the environment and responding to critical events such as natural disasters. Predictive models help predict climate patterns, identify areas of environmental risk and assess the impacts of human activities on the ecosystem. This data is crucial for developing mitigation strategies, adapting to climate change and conserving biodiversity. In summary, Big Data datasets play a fundamental role in environmental studies, providing a comprehensive and real-time understanding of environmental challenges, enabling the implementation of effective strategies for conservation and sustainability.
- Asymptotic statistical results: theory and practicePublication . Kitsos, Christos P.; Oliveira, AmilcarThe target of this paper is to discuss the existent difference of Asymptotic Theory in Statistics comparing to Mathematics. There is a need for a limiting distribution in Statistics, usually the Normal one. Adopting the sequential principle the first-order autoregression model and the stochastic approximation are referred for their particular interest for asymptotic results.
- On the computational methods in non-linear design of experimentsPublication . Kitsos, Christos P.; Oliveira, Amilcar; Kitsos, Christos P.In this paper the non-linear problem is discussed, for point and interval computational estimation. For the interval estimation an adjusted formulation is discussed due to Beale’s measure of non-linearity. The non-linear experimental design problem is regarded when the errors of observations are assumed i.i.d. and normally distributed as usually. The sequential approach is adopted. The average-per-observation information matrix is adopted to the developed theoretical approach. Different applications are discussed and we provide evidence that the sequential approach might be the panacea for solving a non-linear optimal experimental design problem.
- Application of BIBDR in health sciences using RPublication . Oliveira, Amilcar; Oliveira, Teresa A.The role of Experimental Design is very well known, considering applications to a broad range of areas, such as Agriculture, Biology, Medicine, Industry, Education, Economy, Engineering and Food Consumption Sciences. Motivated by the variety of problems faced in the several areas and simultaneously taking advantage of the emerging technological developments, new theoretical results, as well as new designs and structures, have been developed by researchers and practitioners accordingly to the needs. Experimental Design got a place among the most important statistical methodologies and, mainly because of allowing to separate variation sources, since the last century it has been strongly recommended for Health Sciences studies. In this area, particular attention has been devoted to Randomized Complete Block Designs and to Balanced Incomplete Block Designs (BIBD) - which allow testing simultaneously a number of treatments bigger than the block size. Thus, after a brief review of some particular BIBD properties and of BIBDR - Balanced Incomplete Blocks with Block Repetition, an applications to Health Sciences simulated data is illustrated, by exploring R software