Logo do repositório

Matemática e Estatística / Mathematics and Statistics

URI permanente desta comunidade:

http://hdl.handle.net/10400.2/1625

Navegar

Percorrer Matemática e Estatística / Mathematics and Statistics por Objetivos de Desenvolvimento Sustentável (ODS) "13:Ação Climática"

A mostrar 1 - 2 de 2
  • Improving forecasting by resampling STL decomposition
    Publication . Cordeiro, Clara; Ramos, Maria do Rosário; Neves, M. Manuela
    The development of new forecasting algorithms has shown an increasing interest due to the emerging of new fields of application like machine learning and forecasting competitions. Although initially intended for independent random variables, bootstrap methods can be successfully applied to time series. The Boot.EXPOS procedure, which combines bootstrap and exponential smoothing methods, has shown promising results for forecasting. This work proposes a new approach to forecasting, which is briefly described as follows: using Seasonal-Trend decomposition by Loess (STL), the best STL fit is selected by testing all possible combinations of parameters. The best combination of smoothing parameters is chosen based on an accuracy measure. The time series is then decomposed into components according to the best STL fit. The Boot.EXPOS procedure is employed to forecast the seasonal component and the seasonally adjusted time series. These forecasts are aggregated to obtain a final forecast. The performance of this combined forecast is evaluated using real datasets and compared with other established forecasting methods.
    2023-06-30Documento de conferência Acesso restrito Ver mais
  • Spatial and multivariate statistics in assessing water quality in the North Sea
    Publication . Ody, Christopher; Ramos, Maria do Rosário; Carolino, Elisabete Teresa Mata Almeida; Gervasi , O.; Murgante , B.; Garau , C.; Taniar, D.; Rocha , A. M. A. C.; Lago , M. N. Faginas
    The Southern North Sea region plays a vital role for both the economy and society of the surrounding countries. Analyzing the quality of your water is a critical process that involves an assessment of physical, chemical, and biological parameters, essential to guarantee environmental sustainability and the health of local communities and marine ecosystems. Using Multivariate and Spatial Statistics methods, this study seeks to identify spatial patterns and autocorrelations to assess water quality in that region. The data set used was taken on a scientific cruise carried out in December 2020 aboard the RV Meteor vessel, led by a team of German researchers. The raw data went through pretreatment guided by the Data Quality Control protocol of SeaDataNet, an international oceanography project aimed at making European maritime data available. Spike and gradient tests were performed, in addition to data standardization and imputation through inverse distance weighting interpolation. For a better understanding of the scientific area, the data were aggregated by zones for certain analyses, and were sometimes considered globally. An exploratory spatial data analysis (ESDA) was carried out in order to summarize its main characteristics. A reduction in the dimensionality of the original data was carried out through principal component analysis as an auxiliary tool for spatial analysis. The Spatial autocorrelation is analyzed by calculating global and local Moran’s I Statistics. The outcomes indicate a significant spatial autocorrelation for all variables considered in the freshwater areas and a notable range flattening of the variables in the open sea areas, which possibly caused the lack of significant spatial autocorrelation in those areas.
    2024-07Documento de conferência Acesso aberto Ver mais