1,721,032 research outputs found
A weighted approach for spatio-temporal clustering of COVID-19 spread in Italy
No full textThe SARS-Cov-2 has spread differently over space and time worldwide. By monitoring the contagion’s time evolution, the November 3 2020 the Italian government introduced differentiated regime of restrictions among its regions. This experiment demonstrated that public health policies can be effectively designed by means of clustering. This paper proposes a fuzzy clustering model where spatial and temporal dimensions of the disease spread are optimally weighted. The resulting model is applied with the aim of identifying groups of Italian regions with similar contagion spread. We found that two groups of regions sharing similar patterns of COVID-19 spread over both space and time exist. Appropriate public health policies can be designed on the basis of this evidence
Hybrid statistical process monitoring of wire Arc additive manufacturing with frequency informed deep learning
No full textArc welding is classified as a special process under ISO standards, making process monitoring a critical
component of the welding and Additive Manufacturing (AM) certification procedure. Nowadays, the advancements
in data analysis have led to the growing use of Machine Learning (ML) techniques for real-time
weld quality assessment. However, due to their simple design and minimal data requirements, traditional
Statistical Process Monitoring (SPM) methods, such as control charts, remain widely used for evaluating
process quality and detecting anomalies. Despite their significance, traditional SPM techniques struggle
when dealing with multi-variate and high-frequency data typical of Industry 4.0 contexts, making their application
challenging and highlighting the need for new approaches to data analysis. Therefore, in this study,
we propose an innovative hybrid deep learning-based SPM technique for in situ monitoring of theWire Arc
Additive Manufacturing (WAAM) process, with the aim of making SPM more effective in this setting. In
particular, an experimental campaign was conducted using the Invar36 alloy, and an online anomaly detection
application was developed using ML methods to improve the performance of SPM. Specifically, a
Frequency-Informed Convolutional AutoEncoder (FICA) is used as a sensor fusion technique for welding
current and welding voltage data. The obtained latent space across additional temporal dimensions – which
fuse the high-frequency information in a low dimensional space - is then analysed using an Exponentially
Weighted Moving Average (EWMA) chart to detect anomalies during production. The results demonstrate
that the proposed methodology improves anomaly detection performance compared to conventional SPM
techniques, with the F2-score improving from 71.1% to 81.3%
Time series clustering for high-dimensional portfolio selection: a comparative study
No full textIn high-dimensional portfolio selection, traditional asset allocation techniques often yield suboptimal results out-of-sample, while equally weighted portfolios have shown better performances in such scenarios. To leverage the advantages of diversification while addressing the curse of dimensionality, we turn to clustering techniques. Specifically, we explore the application of k-meansclustering for time series, which offers a clear financial interpretation as the prototype of each cluster represents an equally weighted portfolio of the assets within the cluster. In this paper, we conduct a comprehensive comparison of various time series clustering techniques in the context of portfolio performance. By evaluating the out-of-sample performance of portfolios constructed using different clustering approaches, we aim to identify the most effective method for investment purposes
Investors’ attention and network spillover for commodity market forecasting
No full textThis paper explores the role of network spillovers in commodity market forecasting and proposes a novel factoraugmented dynamic network model. We focus on a novel network definition based on investors’ attention to
commodities, positing that commodities exhibit spillovers if they share a similar level of interest. To this
aim, we employ Google Trends search data as an instrumental measure for attention. The results reveal that
including attention-driven spillovers significantly enhances the forecasting accuracy of commodity returns
Kendall correlations and radar charts to include goals for and goals against in soccer rankings
No full textThis paper deals with the challenging themes of the way sporting teams and athletes are ranked in sports competitions. Starting from the paradigmatic case of soccer, we advance a new method for ranking teams in the official national championships through computational statistics methods based on Kendall correlations and radar charts. In detail, we consider the goals for and against the teams in the individual matches as a further source of score assignment beyond the usual win-tie-lose trichotomy. Our approach overcomes some biases in the scoring rules that are currently employed. The methodological proposal is tested over the relevant case of the Italian "Serie A" championships played during 1930-2023
For whom the bell tolls. A spatial analysis of the renewable energy transition determinants in Europe in light of the Russia-Ukraine war
No full textThe ongoing invasion of Russia of Ukraine and energy crises have sparked concern about economic and geopolitical stability all over the world. In Europe, the war has destabilized and endangered the energy cooperation and transition between European countries within and outside of the EU. This emergency has shown once more the importance of energy resilience policies to offset the vulnerability of energy systems and energy insecurity at the national and regional levels. Consilience has been reached on the necessity of enhancing EU energy security as an adaptation strategy. This work contributes to the existing scholarship on renewable energy transition and citizens' perception, focusing on European Union member states. Key characteristics of the renewable energy transition in the EU prior to the energy crisis and the war in Ukraine are examined. To this end, we analyze selected economic, environmental, social, policy and political variables on energy sorting from the Eurobarometer studying European citizens’ perceptions. The exercise makes use of spatially-clustered regression to explore spatial heterogeneity and to elicit determinant information on specific regional groups. We learn that southern Europeans attribute less importance to energy infrastructure to facilitate the renewable energy transition and repute the EU solidity not a requirement for energy security access. Conversely, northern European citizens tend not to associate the responsibility of the EU in guiding competitiveness and policy toward green energy sources transformation. Robustness tests confirm our hypothesis. Regardless of regional differences, the EU energy and ecological transition will thrive with industrial and political cohesion. This process will pass through increased trust in institutions and dedicated energy policy action which will smooth the risks and disruptions coming from current and future shocks
Conditional moments based time series cluster analysis
No full textAbstract In this paper, we propose a new approach for clustering time series showing
similar time-varying moments. At this aim, we compute a dissimilarity measure
assuming that the estimated conditional moments are continuous functions indexed
by time. Conditional moments based clustering allows to obtain different classifications
according to the data distribution’s parameters. We show the usefulness of the
proposed clustering procedure with an application to the financial time series in the
DAX30 inde
Improving out-of-sample forecasts of stock price indexes with forecast reconciliation and clustering
No full textIn this paper, we propose a novel approach to improving forecasts of stock market indexes by considering common stock prices as hierarchical time series, combining clustering with forecast reconciliation. We propose grouping the individual stock price series in various ways including via metadata and using unsupervised learning techniques. The proposed approach is applied to the Dow Jones Industrial Average Index and the Standard & Poor 500 Index and their component stocks, and the results obtained with different grouping approaches are compared. The results empirically demonstrate that the combined use of clustering and reconciliation improves the forecast accuracy of the stock market indexes and their constituents
Conditional moments based time series cluster analysis
No full textAbstract In this paper, we propose a new approach for clustering time series showing
similar time-varying moments. At this aim, we compute a dissimilarity measure
assuming that the estimated conditional moments are continuous functions indexed
by time. Conditional moments based clustering allows to obtain different classifications
according to the data distribution’s parameters. We show the usefulness of the
proposed clustering procedure with an application to the financial time series in the
DAX30 inde
- …
