Francesca Marta Lilja Di Lascio

Missing values in multivariate dependent variables may occur during data collection, requiring imputation methods capable of handling complex inter-variable relationships. We propose a nonparametric copula-based method for imputing dependent multivariate missing data, called NPCoImp. By leveraging the empirical beta copula to calculate the conditional cumulative probability function of the missing variables given the observed ones, NPCoImp imputes data while respecting the underlying distributional shape - particularly radial symmetry - and accordingly adjusts the multivariate values used for imputation. NPCoImp is highly flexible and can handle multivariate missing data with any type of missingness pattern. The performance of NPCoImp has been evaluated through an extensive Monte Carlo study and compared with classical imputation methods, as well as with its direct competitor, the CoImp algorithm, and a machine learning-based technique, the missForest. Our findings indicate that NPCoImp is particularly effective in preserving complex dependence structure irrespective of the sample size, the proportion of missing data and the strength of the dependence. The strong performance of the proposed method is further supported by empirical case studies in the agricultural sector. Finally, the NPCoImp algorithm has been implemented in the R package , which is available on CRAN.

This paper proposes a novel application of a dynamic threshold regression model to three-way panel data, in which the third dimension influences both the identification of regimes and their associated thresholds. Estimation is carried out using the first-difference generalized method of moments with instrumental variables, and a criterion is proposed for the identification of change points in the temporal dynamics. The performance of the proposed methods is assessed through an extensive Monte Carlo simulation study, evaluating the accuracy of both coefficient and threshold estimates, as well as the precision of the detected change points. The methodology is then applied to an innovative three-way panel dataset consisting of fruit bioimpedance measurements collected across a range of frequencies using two types of electrical impedance analyzers: a bench-top device and a state-of-the-art portable one. The analysis provides insights into ripening dynamics by estimating thresholds and identifying change points across the frequency spectrum. In addition, it offers new evidence regarding the performance of the portable device, supporting its practical relevance for post-harvest monitoring, processing, and optimisation within the fruit supply chain. The proposed three-way panel model is implemented in the R package .

Investigating thermal energy demand is crucial for developing sustainable cities and the efficient use of renewable sources. Despite the advances made in this field, the analysis of energy data provided by smart grids is currently a demanding challenge due to their complex multivariate structure and high dimensionality. In this article, we propose a novel copula‐based dissimilarity measure suitable for analyzing district heating demand and introduce a procedure to apply it to high‐temporal resolution panel data. Inspired by the characteristics of the considered data, we explore the usefulness of the Ali‐Mikhail‐Haq copula in defining a new dissimilarity measure to cluster variables in the hierarchical framework. We show that our proposal is particularly sensitive to small dissimilarities based on tiny differences in the strength of the dependence between the involved random variables. Therefore, the measure we introduce is able to distinguish between objects with low dissimilarity better than standard rank‐based dissimilarity measures. Moreover, our proposal considers a weighted version of the copula‐based dissimilarity that embeds the spatial location of the involved objects. We investigate the proposed measure through Monte Carlo studies and compare it with an analogous dissimilarity measure based on Kendall's correlation. Finally, the application to real data concerning the Italian city Bozen‐Bolzano makes it possible to find clusters of buildings homogeneous with respect to their main characteristics, such as energy efficiency and heating surface. In turn, our findings may support the design, expansion, and management of district heating systems.

Clustering methods are widely used in the analysis of gene expression data for their ability to uncover coordinated expression profiles. One important goal of clustering is to discover co–regulated genes because it has been postulated that co–regulation implies a similar function. In the context of agglomerative hierarchical clustering, we introduced a dissimilarity measure based on the Wilks’ Λ statistic that they called the Wilks’ dissimilarity and showed its usefulness in the identification of transcription modules. In this paper, we discuss the ability of the Wilks’ dissimilarity to identify clusters of co-expressed genes by providing an example where the most commonly used dissimilarity measures fail. Furthermore, we carry out a set of simulations aimed to investigate the use of a sparse canonical correlation technique in the estimation of the Wilks’ dissimilarity and provide guidelines for its use.

Investigating thermal energy demand is crucial for the development of sustainable cities and efficient use of renewable sources. Despite the advances made in this field, the analysis of energy data provided by smart grids is currently a demanding challenge. In this paper, we develop a clustering methodology based on a novel dissimilarity measure to analyze a high temporal resolution panel data for district heating demand in the Italian city Bozen-Bolzano. Starting from the characteristics of this data, we explore the usefulness of the Ali-Mikhail-Haq copula in defining a new dissimilarity measure to cluster variables in a hierarchical framework. We show that our proposal is particularly sensitive to small dissimilarities based on tiny differences in the dependence level. Therefore, the proposed measure is able to better distinguish between objects with low dissimilarity than classic rank-based dissimilarity measures. Moreover, our proposal is defined in a spatial version that is able to take into account the spatial location of the compared objects. We investigate the proposed measure through Monte Carlo studies and compare it with the corresponding spatial Kendall's correlation-based dissimilarity measure. Finally, the application to real data makes it possible to find clusters of buildings homogeneous with respect to their main characteristics, such as energy efficiency and heating surface, to support the design, expansion and management of district heating systems.

A theoretical framework is presented for a (copula-based) notion of dissimilarity between continuous random vectors and its main properties are studied. The proposed dissimilarity assigns the smallest value to a pair of random vectors that are comonotonic. Various properties of this dissimilarity are studied, with special attention to those that are prone to the hierarchical agglomerative methods, such as reducibility. Some insights are provided for the use of such a measure in clustering algorithms and a simulation study is presented. Real case studies illustrate the main features of the whole methodology.