Publications 2021


16.  Estatística: desafios transversais às ciências com dados: atas do XXIV Congresso da Sociedade Portuguesa de Estatística

Milheiro, Paula and Pacheco, António and Sousa, Bruno de and Alves, Isabel Fraga and Pereira, Isabel and Polidoro, Maria João and Ramos, Sandra

Sociedade Portuguesa de Estatística

Sem resumo disponível.

Book Chapters

15.  Temperature time series forecasting in The Optimal Challenges in Irrigation (TO CHAIR)

Gonçalves, A. Manuela and Costa, Cláudia and Costa, Marco and Lopes, Sofia O. and Pereira, Rui

Advances in Evolutionary and Deterministic Methods for Design, Optimization and Control in Engineering and Sciences. Computational Methods in Applied Sciences


Predicting and forecasting weather time series has always been a difficult field of research analysis with a very slow progress rate over the years. The main challenge in this project—The Optimal Challenges in Irrigation (TO CHAIR)—is to study how to manage irrigation problems as an optimal control problem: the daily irrigation problem of minimizing water consumption. For that it is necessary to estimate and forecast weather variables in real time in each monitoring area of irrigation. These time series present strong trends and high-frequency seasonality. How to best model and forecast these patterns has been a long-standing issue in time series analysis. This study presents a comparison of the forecasting performance of TBATS (Trigonometric Seasonal, Box-Cox Transformation, ARMA errors, Trend and Seasonal Components) and regression with correlated errors models. These methods are chosen due to their ability to model trend and seasonal fluctuations present in weather data, particularly in dealing with time series with complex seasonal patterns (multiple seasonal patterns). The forecasting performance is demonstrated through a case study of weather time series: minimum air temperature. | doi | Peer Reviewed

14.  Análise de regressão linear com autocorrelação nos erros para dados censurados

Sousa, Rodney and Pereira, Isabel and Silva, Maria Eduarda

Estatística: desafios transversais às ciências com dados: atas do XXIV Congresso da Sociedade Portuguesa de Estatística

Sociedade Portuguesa de Estatística

Este trabalho aborda, numa perspetiva bayesiana, a análise de modelos de regressão linear com erros autocorrelacionados para dados censurados, recorrendo a métodos Computacionais Bayesianos Aproximados (ABC) e ao amostrador de Gibbs com a Ampliação de Dados (GDA). Considera-se que o termo dos erros segue um processo autorregressivo, AR, e investiga-se o desempenho dos métodos através de dois estudos de simulação com diferentes cenários de censura (5%, 20% e 40%) e dimensão de amostras (50, 100 e 500). Os resultados indicam que o método GDA é consistente Bayesiano, mesmo em cenários em que a proporção de valores censurados é elevada, enquanto que no método ABC, as estimativas dependem fortemente das distribuições a priori. | Peer Reviewed

13.  Improving Short-term Forecasts of Daily Maximum Temperature with the Kalman Filter with GMM Estimation

Costa, Marco and Pereira, Fernanda Catarina and Gonçalves, A. Manuela

The 21st International Conference on Computational Science and Applications (ICCSA 2021)

Springer, Cham

Within the scope of the TO CHAIR project, a state space modeling approach is proposed in order to improve accuracy obtained from the website with a dataset of real observations. The proposed model establishes a stochastic linear relationship between the maximum temperature observed and the h-step-ahead forecast pro- duced from the website. This relation is modeled in a state space frame- work associated to the Kalman filter predictors. Since normality of dis- turbances was not a good assumption for this dataset, alternative Gen- eralized Method of Moments (GMM) estimators were considered in the models parameters estimation. The results show that this approach al- lows reducing the RMSE of the uncorrected forecasts in 16.90% consider- ing the 6-step-ahead forecasts and in 60.45% considering the 1-step-ahead forecasts, compared with the initial RMSE. Additionally, empirical con- fidence intervals at the 95% level have a coverage rate similar to this confidence level. So, this approach has proven suitable for this type of forecasts correction since it considers a stochastic calibration factor in order to model time correlation of this type of variable. | doi | Peer Reviewed

12.  Change point detection in a state space framework applied to climate change in Europe

Monteiro, Magda and Costa, Marco

Computational Science and Its Applications: ICCSA 2021


This work presents the statistical analysis of time series of monthly average temperatures in several European locations using a state space approach, where it is considered a model with a deterministic seasonal component and a stochastic trend. Temperature rise rates in Europe seem to have increased in the last decades when compared with longer periods, hence change point detection methods were applied to residuals state space models in order to identify these possible changes in the monthly temperature rise rates. In Northern Europe the change points were, almost all, identified in the late 1980s while in Central and Southeastern Europe was, for the majority of cities, in the 1990s and later. | doi | Peer Reviewed

11.  CDPCA: 10 years after

Freitas, Adelaide

Estatística: desafios transversais às ciências com dados: atas do XXIV Congresso da Sociedade Portuguesa de Estatística

Sociedade Portuguesa de Estatística

Clustering and Disjoint Principal Component Analysis (CDPCA) is a constrained principal component analysis for multivariate numerical data. The main goal is to detect clusters of objects and, simultaneously, to fi nd a partitioning of variables such that the between cluster deviance in the reduced space of such partition is maximized. The partition formed by a disjoint set of the original variables identifi es the groups of variables belonging to the CDPCA components. Recently, this methodology has been implemented in a R-function called CDpca. In this work, we review some theoretical issues of the CDPCA model and present two applications on real data sets using the R-function CDpca. | Peer Reviewed

10.  Multivariate sustainability profile of Global Fortune 500 companies using GRI-G4 database

Jiménez-Hernández, Mónica and Vicente-Galindo, Purificación and Tejedor-Flores, Nathalia and Freitas, Adelaide and Galindo, Purificación

Handbook of Research on Applied Data Science and Artificial Intelligence in Business and Industry

IGI Global

The main objective of this research is to find the sustainability gradients of Global Fortune 500 companies and sort them as a function of economic, environmental, and social components using multivariate statistical methods to establish the foundations for better knowledge of the trends and sustainability reporting habits. A combined approach, comprising principal coordinates analysis (PCoA) and logistic regression model (LRM), is proposed to build an external logistics biplot (ELB). Moreover, HJ-Biplot and parallel coordinates are applied. This chapter helps to understand why many companies view their corporate social responsibility (CSR) reports as a way to guarantee the credibility of the published information. In particular, based on the Global Reporting Initiative, the sustainability gradients of the Global Fortune 500 companies are obtained and statistically exploited to analyze how the companies can make improvements in terms of sustainability. | doi | Peer Reviewed

9.  Math requirements for admission in elementary school teacher education programs: does it matter?

Hall, A. and Alvelos, H. and Xambre, A. R. and Hall, F. and Costa, A. T. and Silva, P.

EDULEARN21 Proceedings


In Portugal, the initial training of teachers for the first three levels of education (children from 3 to 12 years old) is carried out through a single 3-years undergraduate degree, called Basic Education. This training is completed with a professional master's degree where future teachers choose which levels they will teach. Most master's degrees cover two levels of education: pre-school and 1st cycle or 1st and 2nd cycles. This means that most teachers become qualified to teach the 1st cycle, usually known as elementary school. First cycle teachers teach all educational areas, including mathematics. It is, therefore, essential that they finish their degree in Basic Education with a solid background in Mathematics. Until the academic year 2017-18 there was no mandatory entry requirement in Mathematics to the Basic Education Degree and many students were admitted without having had any Mathematics in secondary education. To ensure a better pre-university training in Mathematics, the Portuguese government imposed an admission exam, as from the academic year 2018-19. This work aims to assess the impact of the government measure on the training of future teachers in the area of mathematics, through the analysis of the performance of students in the 1st year of the Basic Education Degree, from a Portuguese university, in a mathematics course of the 1st semester, over the academic years 2017-18 to 2019-20. The statistical study carried out reveals that there were significant improvements in the results of the students. An analysis of the gender tendency in the choice of the teaching profession was also done and showed that there is a disproportion between men and women in pre-service teachers, since most young people who want to become teachers are female. This study contributes to a better understanding of the impact certain measures and policies can have on the quality of Higher Education academic training. This type of approach can be applied to other similar situations, for other programs and other courses, helping, in this way, decision making in Higher Education admission policies, as well as researchers in this field. | doi | Peer Reviewed


8.  Bivariate models for time series of counts: a comparison study between PBINAR models and dynamic factor models

Monteiro, Magda and Pereira, Isabel and Scotto, Manuel G.

Communications in Statistics - Simulation and Computation

Taylor and Francis

The aim of this work is to assess the modeling performance of two bivariate models for time series of counts, within the context of a forest fires analysis in two counties of Portugal. The first model is a periodic bivariate integer-valued autoregressive (PBINAR), easily interpreted due to the PINAR description of each component. The alternative model is a bivariate dynamic factor (BDF) that has a flexible structure, with the dynamics described through the mean value of each component that is a function of latent factors. The results reveal that BDF model exhibits a better ability to capture the dependence structure. | doi | Peer Reviewed

7.  On the theory of periodic multivariate INAR processes

Santos, Cláudia and Pereira, Isabel and Scotto, Manuel G.

Statistical Papers


In this paper a multivariate integer-valued autoregressive model of order one with periodic time-varying parameters, and driven by a periodic innovations sequence of independent random vectors is introduced and studied in detail. Emphasis is placed on models with periodic multivariate negative binomial innovations. Basic probabilistic and statistical properties of the novel model are discussed. Aiming to reduce computational burden arising from the use of the conditional maximum likelihood method, a composite likelihood-based approach is adopted. The performance of such method is compared with that of some traditional competitors, namely moment estimators and conditional maximum likelihood estimators. Forecasting is also addressed. Furthermore, an application to a real data set concerning the monthly number of fires in three counties in Portugal is presented. | doi | Peer Reviewed

6.  Language assessment in awake brain surgery: the Portuguese adaptation of the Dutch linguistic intraoperative protocol (DuLIP)

Alves, Joana and Cardoso, Mafalda and Morgado, Mariana and De Witte, Elke and Satoer, Djaina and Hall, Andreia and Jesus, Luis M. T.

Clinical Linguistics & Phonetics

Taylor & Francis

Awake brain surgery, combined with neurophysiological evaluation and intraoperative mapping, is one of the preferential lines of treatment when approaching low-grade gliomas. Speech and language assessment is used while applying Direct Electrical Stimulation (DES) and during the resection of a lesion/tumour, as it allows to establish related eloquent areas and optimise the extent of the resection and avoid impairments. Patients need to be assessed pre, intra and post-surgery, but in under resourced countries such as Portugal, there are still no standardised and validated tools to conduct this type of evaluation. To address this need, the tasks of the Dutch Linguistic Intraoperative Protocol (DuLIP) were adapted to European Portuguese, and the resulting materials were standardised for a group of 144 Portuguese participants. For each task, the impact of age, gender and schooling were measured. The resulting Portuguese version of the DuLIP (DuLIP-EP) consists of 17 tasks, including phonological, syntactic, semantic, naming and articulatory tests. No significant differences were found between male and female participants. However, schooling influenced phonological and syntactic fluency, object naming and verb generation. Schooling and age had a significant impact on semantic fluency and reading with semantic odd word out tasks. This is the first contribution to the standardisation of a tool that can be used during an awake brain surgery in Portugal, which includes a new phonological odd word out task that is not currently available in the Dutch version. | doi | Peer Reviewed

5.  Investigating carbon emissions from electricity generation and GDP nexus using maximum entropy bootstrap: evidence from oil-producing countries in the Middle East

Zanjani, Zeinab and Macedo, Pedro and Soares, Isabel



The maximum entropy bootstrap for time series is applied in this study to investigate the nexus between carbon emissions from electricity generation and the gross domestic product, using a bivariate framework for eight Middle Eastern countries between 1995 and 2017. The sample under study includes oil-producing countries such as Bahrain, Iran, Iraq, Kuwait, Oman, Qatar, Saudi Arabia, and United Arab Emirates. As the electricity generation in these economies relies mainly on oil and gas, finding out the existence and direction of the relationship between the two considered variables has remarkable implications for policymakers and governments in these countries to achieve both higher economic growth and environmental protection. As expected, this nexus is validated for all countries in the sample but not in all models, time periods, and lags. Therefore, policymakers can set appropriate electricity conservation policies based on these varied empirical findings to boost economic growth with minimum environmental degradation. | doi | Peer Reviewed

4.  An empirical comparison of two approaches for CDPCA in high-dimensional data

Freitas, Adelaide and Macedo, Eloísa and Vichi, Maurizio

Statistical Methods & Applications


Modifed principal component analysis techniques, specially those yielding sparse solutions, are attractive due to its usefulness for interpretation purposes, in particular, in high-dimensional data sets. Clustering and disjoint principal component analysis (CDPCA) is a constrained PCA that promotes sparsity in the loadings matrix. In particular, CDPCA seeks to describe the data in terms of disjoint (and possibly sparse) components and has, simultaneously, the particularity of identifying clusters of objects. Based on simulated and real gene expression data sets where the number of variables is higher than the number of the objects, we empirically compare the performance of two diferent heuristic iterative procedures, namely ALS and two step-SDP algorithms proposed in the specialized literature to perform CDPCA. To avoid possible efect of diferent variance values among the original variables, all the data was standardized. Although both procedures perform well, numerical tests highlight two main features that distinguish their performance, in particular related to the two-step-SDP algorithm: it provides faster results than ALS and, since it employs a clustering procedure (k-means) on the variables, outperforms ALS algo rithm in recovering the true variable partitioning unveiled by the generated data sets. Overall, both procedures produce satisfactory results in terms of solution precision, where ALS performs better, and in recovering the true object clusters, in which two-step-SDP outperforms ALS approach for data sets with lower sample size and more structure complexity (i.e., error level in the CDPCA model). The proportion of explained variance by the components estimated by both algorithms is affected by the data structure complexity (higher error level, the lower variance) and presents similar values for the two algorithms, except for data sets with two object clusters where the two-step-SDP approach yields higher variance. Moreover, experimental tests suggest that the two-step-SDP approach, in general, presents more ability to recover the true number of object clusters, while the ALS algorithm is better in terms of quality of object clustering with more homogeneous, compact and well separated clusters in the reduced space of the CDPCA components. | doi | Peer Reviewed

3.  Using an interdisciplinary approach to the teaching of solid geometry in a professional development course for preschool and primary school teachers

Hall, Andreia and Pais, Sónia

Indagatio Didactica


This paper presents some results of a professional development course for teachers where the participants studied basic solid geometry and developed applied projects in an interdisciplinary context. The course took place in a Portuguese university, from February to May 2020, and involved 19 teachers of preschool and primary levels (grades 1 to 4). The authors have developed a qualitative case study to evaluate how an interdisciplinary approach to the teaching of solid geometry is perceived, by the mathematics teachers, as a contribution to the teaching/learning process of geometry. Overall, the activities developed have proved to be successful examples of interdisciplinary methodologies. Moreover, the approach followed during the course helped the teachers develop their geometric competences concerning solid geometry in a more consistent appropriation and application of the geometric concepts involved. | doi | Peer Reviewed

2.  Mathematics classes for tourism undergraduate students and pre-service teachers with active methodologies using technologies

Santos, Vanda and Pais, Sónia and Hall, Andreia

International Journal of Technology in Mathematics Education

Research Information

In the last few decades, technology has advanced in multiple fields, including Education. Some of its benefits include improving student performance and motivation, fostering active learning and tracking student progress. Game-based learning platforms, like Kahoot!, can be used for reviewing content and motivating students for learning. The participants in the study are undergraduate and post-graduate students from two Portuguese public higher education institutions. The aim of the study is to investigate students’ perceptions of how Kahoot! can be used as a tool for reviewing class content or designing warm-up activities. A quantitative survey is being conducted to gather information about students’ insights on the use of Kahoot!. Other studies have also shown that higher education students are usually receptive to the use of this tool, finding it useful to increase their motivation and considering that technology can positively impact learning. | doi | Peer Reviewed


1.  Behaviour of the one-minute sit-to-stand test during six months in people with COPD

Cabral, Jorge and Afreixo, Vera and Rocha, Vânia and Souto-Miranda, Sara and Marques, Alda

Journal of Statistics on Health Decision

Universidade de Aveiro; Centro Hospitalar do Baixo Vouga

No abstract available. | doi | Peer Reviewed
(latest changes on 2021-12-02 10:09)

© 2019 CIDMA, all rights reserved