Abstract: The increasing need to analyze large and complex datasets requires the development of innovative statistical methodologies. Traditional data tables, in which each cell contains a single quantitative or categorical value, are often insufficient to represent the variability inherent in many contemporary data structures. Within the framework of Symbolic Data Analysis (SDA), distributional data are studied, in which each individual corresponds to an empirical distribution. The development of statistical models and methods to represent, analyze, interpret, and structure distributional data has increased significantly in recent years. In this context, linear models provide a fundamental methodological framework, supporting techniques such as linear regression, linear discriminant analysis, and principal component analysis, enabling the effective analysis of complex data structures.
Sónia Dias is an Adjunct Professor in the Department of Mathematics at ESTG – IPVC and a researcher at LIAAD, INESC TEC, University of Porto (UP). Her main research interests include Data Analysis, Symbolic Data Analysis (analysis of complex multidimensional data), and Statistical/Mathematical Applications. In recent years, she has presented several communications at national and international conferences and has published articles in international journals and conference proceedings. She is also co-editor of the book Analysis of Distributional Data, published in 2022. Since 2020, she has been a member of the Board of CLAD – Portuguese Association for Classification and Data Analysis, and since 2023, secretary of IFCS – International Federation of Classification Societies.
ORCID: 0000-0002-2100-2844