Abstract: The paper investigates the deployment of data analytics and machine learning to improve welding quality in Tecnomulipast srl, a small-to-medium sized manufacturing firm located in Puglia, Italy. The firm produces food machine components and more recently mechanized its laser welding process with the introduction of an IoT-enabled system integrating photographic control. The investment, underwritten by the Apulia Region under PIA (Programmi Integrati di Agevolazione) allowed Tecnomulipast to not only mechanize its production line but also embark upon wider digital transformation. This involved the creation of internal data analytics infrastructures that have the capability to underpin machine learning and artificial intelligence applications. This paper addresses a prediction of weld bead width (LC) with a dataset of 1,000 observations. Input variables are laser power (PL), pulse time (DI), frequency (FI), beam diameter (DF), focal position (PF), travel speed (VE), trajectory accuracy (TR), laser angle (AN), gas flow (FG), gas purity (PG), ambient temperature (TE), and penetration depth (PE). The parameters were exploited to build and validate some supervised machine learning algorithms like Decision Trees, Random Forest, K-Nearest Neighbors, Support Vector Machines, Neural Networks, and Linear Regression. The performance of the models was measured by MSE, RMSE, MAE, MAPE, and R². Ensemble methods like Random Forest and Boosting performed the highest. Feature importance analysis determined that laser power, gas flow, and trajectory accuracy are the key variables. This project showcases the manner in which Tecnomulipast has benefited from public investment to introduce digital transformation and adopt data-driven strategies within Industry 4.0.

Abstract: This case study features Tecnomulipast, an SME from Southern Italy that specializes in machinery production for the food processing industry. The study is in fact centered on the company's digital transformation process, facilitated by investments in advanced production systems and innovation-driven managerial practices both facilitated by regional co-financing initiatives, including from Regione Puglia. At the center of it all is the integration between a new Industry 4.0-compliant laser welding system in the company's ERP system. Through Internet of Things (IoT) technologies, the system is inherently equipped to collect and transmit batch-level as well as real-time data, instantiating a cyber-physical system for advanced manufacturing. Easy to connect by standard interface (i.e. OPC-UA), the system is tied to an analytics data framework capable of working on structured data (e.g., KPIs, sensors' metrics) as well as on unstructured data (e.g., images), allowing for real-time monitoring, early anomaly signaling, and optimization of processes. Designed for scalability, the related technology architecture is future-proof to include artificial intelligence (AI) integration for augmenting decision-making with predictive and prescriptive analytics. Beyond the technological enhancement, however, the transformation was facilitated by an excellence managerial model that focuses on flexibility, data-driven governance, as well as on constant learning. Tecnomulipast's case offers an replicable template for SMEs—especially in low digital maturity areas—showing that targeted investment, innovation-driven management, and system-level integration might finally eliminate the gap between tech potential and operational performance in Industry 4.0 transitions.

Abstract: This paper examines the impact of climate risks on urban economic resilience, using panel data from Chinese cities between 2009 and 2022. A multidimensional indicator system, encompassing recovery resilience, adaptive resilience, and transformative innovation capacity, is developed to identify how climate risks negatively affect urban economic resilience. The results show that climate risks weaken resilience by reducing population size and destabilizing financial systems. Additionally, these risks have significant spatial spillover effects, extending from local areas to neighboring cities through regional economic networks, with particular impact on geographically adjacent cities. Heterogeneity analysis indicates that developed eastern cities, central cities, and resource-based cities are more vulnerable to climate risks, while cities in central and western regions and non-central cities show greater resilience. The paper proposes policy recommendations to strengthen urban resilience, including investment in climate-adaptive infrastructure, promoting economic diversification, establishing cross-regional climate risk management, developing green finance systems, and raising public awareness of climate risks.

Abstract: Accurate prediction of gold prices is essential for informed financial decision-making, given their sensitivity to economic, political, and social factors. This study proposes a new hybrid framework for forecasting gold prices, combining traditional financial modelling, classical machine learning, and advanced deep learning methods, including long short-term memory networks and their variations. The framework integrates financial, macroeconomic, and sentiment indicators through feature fusion, capturing complex temporal dynamics and cross‑variable dependencies to improve prediction accuracy. The experimental evaluation spans a ten-year period (2014–2024), allowing for a robust assessment of framework performance in forecasting gold futures under varying market conditions. Comparative analysis of classical econometric and modern machine‑learning models shows that advanced methods achieve higher forecasting accuracy and remain robust despite data variability.

Abstract: This research seeks to evaluate the effects of the preceding cyclical indicators and macroeconomic variables on the performance of the Brazilian stock market. The objective is to identify how these factors influence the behavior of the main index representing this market. In this way, we analyze how shocks in the Composite Leading Indicator of the Economy, as well as the Basic Interest Rate of the Economy, the Broad National Consumer Price Index, the Express Nominal Exchange Rate and the Central Bank Economic Activity Index impact the performance of IBOVESPA index. The results obtained indicated that the shocks to the Composite Leading Indicator of the Economy, Exchange Rate and Inflation variables influenced the IBOVESPA in different and statistically significant ways. However, shocks to the Economic Activity Index and the Rate of Interest did not exert a statistically significant influence on the index.

Abstract: The increasing pressure for transparency in corporate sustainability reporting, especially under frameworks such as the Corporate Sustainability Reporting Directive and the European Sustainability Reporting Standards, has raised the need for sector-specific models to integrate financial, social, and environmental indicators coherently and measurably. This study proposes a composite econometric model to assess the sustainability performance of companies in the construction sector in a digital context, a domain that remains underexplored despite its substantial economic and environmental impact. Drawing on a sample of 1,600 Romanian construction companies over a ten-year period (2013–2023), the study develops a multidimensional sustainability score and tests its financial drivers using Ordinary Least Squares regression models. The model incorporates nine financial structure variables as predictors of sustainability outcomes across three dimensions - financial, social, and environmental - while ensuring robustness through heteroscedasticity and multicollinearity diagnostics. Results show that indicators such as return on assets, debt ratio, and equity structure significantly influence sustainability performance, particularly in the financial and environmental dimensions. In contrast, the social dimension exhibits lower explanatory power. The findings suggest that financial resilience plays a critical role in shaping sustainable practices in the construction industry and support the adoption of integrated models for performance benchmarking and policy alignment.

Abstract: Against the backdrop of globalization, supply chain concentration has emerged as a critical factor influencing corporate financial strategies. This study delves into how supply chain concentration, encompassing supplier and customer concentration, affects corporate cash holdings and explores the moderating role of corporate governance. Using data from Chinese A-share listed companies from 2012 to 2023, it employs a fixed-effects ordinary least squares regression model. Interaction terms of board size, the proportion of independent directors, and supply chain concentration are constructed for heterogeneity analysis. The Herfindahl-Hirschman Index is utilized as an instrumental variable to address endogeneity, accompanied by robustness tests. The findings reveal a significant positive correlation between supply chain concentration and cash holdings. Supplier concentration (with a coefficient of 0.0162) and customer concentration (0.0152) both prompt firms to hold higher cash reserves. Corporate governance moderates this relationship: larger boards amplify the effect, while independent directors have no significant influence. This study identifies supply chain concentration as a key liquidity driver, facilitating the integration of supply chain management and corporate finance theories. Practically, it advises enterprises to balance supply chain relationships and governance structures to optimize cash reserves and enhance financial resilience in dynamic markets.

Abstract: This study investigates the relationship between Research Intensity (RI) and a range of Environmental, Social, and Governance (ESG) variables for Italian regions using machine learning algorithms and panel data models. The study seeks to identify the most predictive variables of research intensity from a range of cultural, environmental, socio-economic, and governance indicators. Support Vector Machine, Random Forest, k-Nearest Neighbors, and Neural Network algorithms are used to ascertain comparative predictive power. Feature importance analysis identifies education levels, in particular tertiary education qualifications, and technological infrastructure as most predictive of research intensity. Regional differences in research intensity are also investigated on the basis of political representation, healthcare accessibility, material consumption, and cultural investment variables. Results indicate that economically developed regions with sufficient research capacity are more research-intensive but can also face environmental sustainability and social inclusiveness issues. The study concludes that policy measures to enable education, technological innovation, environmental management, and governance improvement are required to spur research capacity in Italian regions. The study also provides insight into the use of research intensity in informing broader ESG objectives, including policy intervention for mitigating regional imbalances. Future studies should provide insight into the dynamic interaction effects of research intensity and ESG variables over time using more sophisticated machine learning techniques to further enhance predictive power.

Abstract:

Child sex ratios in India have fluctuated over time, with northern states exhibiting greater disparities compared to eastern and southern regions. However, despite policy efforts in the past, early childhood gender ratios are slowly declining in Tamil Nadu, one of the states with relatively equalized child sex ratios. It is imperative that the gender disparity in infant demographics increases over the next few decades given that there is betterment in factors like medical infrastructure and availability of tap water in most villages in Tamil Nadu. Procedure: Data was taken from the Development Data Labs’ SHRUG database, and information on various variables, location, spatial stats, and shapefiles were used to assess how trends in child sex ratio across various districts in Tamil Nadu are affected by literacy rates, worker population, scheduled castes and scheduled tribes. Results: Although the availability of medical infrastructure and tap water does not necessarily affect trends in child sex ratio on a district level in Tamil Nadu, OLS estimates show that there is a significant relationship between literacy rates and worker population on child sex ratio. Conclusion: Policies need to be shaped around quality education and employment opportunities to ensure child sex ratios in Tamil Nadu are equalized in the coming decades through a multidimensional approach—encompassing economic, educational, and social strategies.

Abstract: Sequential recommendation is an extensively explored approach to capturing users’ evolving preferences based on past interactions, aimed at predicting their next likely choice. Despite significant advancements in this domain, including methods based on RNNs and self-attention, challenges like limited supervised signals and noisy data caused by unintentional clicks persist. To address these challenges, some studies have incorporated unsupervised learning by leveraging local item contexts within individual sequences. However, these methods often overlook the intricate associations between items across multiple sequences and are susceptible to noise in item co-occurrence patterns. In this context, we introduce a novel framework, Global Unsupervised Data-Augmentation (UDA4SR), which adopts a graph contrastive learning perspective to generate more robust item embeddings for sequential recommendation. Our approach begins by integrating Generative Adversarial Networks (GANs) for data augmentation, which serves as the first step to enhance the diversity and richness of the training data. Then, we build a Global Item Relationship Graph (GIG) based on all user interaction sequences. Subsequently, we employ graph contrastive learning on the refined graph to enhance item embeddings by capturing complex global associations. To model users’ dynamic and diverse interests more effectively, we enhance the CapsNet module with a novel target-attention mechanism. Extensive experiments show that UDA4SR significantly outperforms state-of-the-art approaches.

Abstract: This paper provides an in-depth examination of Exploratory Spatial Data Analysis (ESDA), an extension of the exploratory data analysis concepts introduced by John Tukey in the 1970s. ESDA is designed to uncover hidden patterns and relationships in spatial data that are not apparent through traditional statistical methods. It incorporates techniques such as spatial visualization, dynamic linking, and spatial autocorrelation analysis to investigate spatial distributions and relationships across different datasets. By employing maps, histograms, scatter plots, and other graphical representations, ESDA enables the interactive exploration of spatial dependencies and heterogeneities. The integration of geographic information systems (GIS) tools like Luc Anselin’s GeoDa software enhances the ability of ESDA to handle complex spatial data structures and to visualize geographical data effectively. This paper discusses various ESDA methodologies tailored to analyze continuous, discrete, and spatial-temporal data, emphasizing their critical role in the development of econometric models. Additionally, it underscores the importance of spatial context in statistical analysis, advocating for the consideration of both physical and socio-cultural dimensions of space in understanding patterns and processes that influence human behavior and environmental dynamics. The detailed review also covers the use of ESDA in multiple disciplines, illustrating its versatility and effectiveness in providing insightful analyses that inform better decision-making in fields ranging from urban planning to environmental science.

Abstract: Poor air quality in India has sparked our interest in studying the time series dynamics of PM2.5 in India's five most populous cities (Mumbai, New Delhi, Hyderabad, Chennai, and Kolkata). Daily data for the period 2014-2023 are examined in the paper. Using fractional integration methods, we analyze the persistence, seasonality, and time trends of the data. The results indicate that all series display fractional degrees of integration, being smaller than 1 and thus presenting mean reversion. Moreover, only for New Dehli and Kolkata the time trends are significantly negative, implying a continuous reduction in the level of pollution. These findings suggest that targeted interventions, such as stricter emission regulations, improved urban planning, and the promotion of clean technologies, are essential to sustain and amplify the observed improvements in air quality. The study also highlights the need for consistent and long-term efforts to address pollution in Mumbai, Hyderabad, and Chennai, where no significant reductions have been observed, emphasizing the importance of adapting policies to regional conditions. The paper's findings can serve as a guide for air pollution management and for policymakers at the Central Pollution Control Board (CPCB) the governmental body responsible for monitoring and regulating environmental pollution in India.

Abstract: This study examines the drivers of inflation in Angola over the period 2015–2024 (120 observations), using an autoregressive distributed lag (ARDL) model. By integrating key macroeconomic variables such as the consumer price index (CPI), trade-weighted inflation (TCPI), nominal exchange rate (EXR), and money supply (M2), the research investigates both short-run dynamics and long-run equilibrium relationships. The analysis applies unit root tests and ARDL bounds testing to confirm cointegration among the variables and employs Granger causality tests to explore directional influences. Results indicate that inflation in Angola exhibits strong persistence with significant adjustment toward long-run equilibrium, while external shocks, particularly exchange rate fluctuations, play a critical role in influencing price levels. Monetary policy transmission and fiscal discipline appear to have limited immediate impact, suggesting that structural factors and reliance on oil revenues are major contributors to inflationary pressures. Based on these findings, the study recommends improvements in foreign exchange management, enhancement of domestic production capacity, diversification of the economy, and tighter fiscal controls.

Abstract: This study adopts the affine term structure three-factor models outlined by \cite{dai2000specification}, aiming to analyse South African (SA) government bond yields across various maturities. The primary objective is to evaluate whether these models offer robust pricing capabilities—being both admissible and flexible—while capturing the conditional correlations and volatilities of yield factors specific to SA bond yields. For a model to be considered admissible, it must also demonstrate economic identification and maximal flexibility. We thus investigate the short-, medium-, and long-term dynamics of bond yields concurrently. Model estimation involves deriving joint conditional densities through the inversion of the Fourier transform applied to the characteristic function of the state variables. This enables the use of maximum likelihood estimation as an efficient method. We assume that the market prices of risk are proportional to the volatilities of the state variables. The analysis reveals negative correlations between factors. Among the models tested, the $A_1(3)$ model outperforms the $A_2(3)$ model in terms of fit, both in-sample and out-of-sample.

Abstract: In this paper, we introduce the Bayesian Gibbs Slice Sampler (BGSS), a novel MCMC algorithm inspired in the Latent Slice Sampling (LSS) framework, where Bayesian inference is employed to refine the proposal distribution required to accommodate the single adjustment parameter. Unlike methods based on gradient calculations or those requiring complex, hard-to-optimize adaptive proposals, BGSS naturally incorporates Bayes’ theorem during the chain adaptation phase to learn about the target distribution. Subsequently, it generates nearly independent proposals derived from a conditionally univariate factorization of the parameter space, along with a QR decomposition, thus conferring substantial efficiency to the exploration process. The proposed sampler is both adaptable and computationally effective, matching the speed of LSS and delivering results on par with state-of-the-art approaches like the No-U-Turn Sampler (NUTS). We display its capabilities through simulated and real-world applications, highlighting an analysis of sovereign credit ratings and illustrating how BGSS can model the influence of macroeconomic fundamentals over multiple time horizons. Overall, BGSS strikes a favourable balance between performance and computational demands, making it a dependable tool for Bayesian inference in econometric contexts.

Abstract: This paper focuses on the econometric methodology developed by Professor David Hendry with his associates along several decades. The paper comments on the statistical foundations of the methodology, which are based on the probability approach in Econometrics introduced by Haavelmo. The paper proposes eleven main features to sum up the methodology and they are discussed with detailed. A pivotal point in the methodology is the, Local Data generation Process (LDGP), which is unknown at the beginning and to discover it, it must be nested in a suitable General Unrestricted Model (GUM). The GUM must include variables from possibly relevant economic theories and all other types of variables that may be necessary to represent the economic system under study, including the indicator variables employed in the Indicator Saturation Estimation (ISE). Professor Hendry invented the ISE to capture as “many contaminating influences as possible” affecting the data. From the GUM a reduction process to discover a final congruent model is carried out by the procedure from general-to-specific. Usually, there are more variables than observations and a sophisticated multiple-path search using segmentation by blocks has been designed. All the process is automated by a machine-learning program, Autometrics. This methodology has also been incorporated into Climate Econometrics.

Abstract:

Board size is an important component of corporate governance. This study selects data from 55 listed companies in the automotive industry with beta values close to 1 from 2012 to 2024 for empirical research, analyzing the relationship between board size and corporate performance, as well as the main factors influencing board size. The results show that board size is positively correlated with corporate performance, and board size is mainly influenced by environmental characteristics, firm size, development stage, and internal board structure. Based on empirical data, the maximum board size is calculated to be 11 members.

Abstract: This article examines the progress of research in the field of "Artificial Intelligence and Food Security" from the beginning to February 11, 2025. The main research method used in this article is bibliometric analysis using VosViewer software, which examines the characteristics of published articles, such as authors, countries, etc. In the study, 497 articles were indexed in the Scopus database from 2009 to February 11, 2025. The findings show that by 2025, the publication of articles is ascending, and the highest number of articles is related to 2024. India has been the largest producer of scientific articles. The findings show that research in the field of artificial intelligence and food security is expanding in various fields. Using artificial intelligence technology, stakeholders will be able to create more efficient and durable food systems that will not only support food security for all but also help protect natural resources for future generations.

Abstract: This study investigates the Industrial Electricity Pricing (IEP) profiles of 22 OECD countries to understand the effect of taxes to overall prices. Clustering analysis was performed on pricing data from the year 2000 to 2018 to observe how prices evolved. Ordinal logit regression analysis was performed to determine possible associations between the clustered groups and the percentage share of renewables generated (REG). Other independent variables indicating economic and market structures were also considered. Clustering results for both prices before and after tax indicated three pricing clusters, termed low, median, and high pricing clusters. IEP in Italy and Germany were found to have the highest effect owing to taxes while IEP in countries such as the US, Norway, Canada, and Denmark were least affected by taxes. Regression results show positive associations between the clustered profiles and REG. The positive association between the non-taxed component of IEP and a unit increase of REG is 1.41 times, whereas the positive association of overall IEP price (including taxes) and a unit increase of REG is 56.26 times, which is 39.9 times higher. Our results show that REG penetration as such has had a minimal effect on IEP over the time under consideration, but rather taxation on IEP coincidental with REG penetration, has contributed to IEP increases.

Abstract: This study proposes an approach to assessing the Value of Statistical Life (VSL) based on the contingent valuation of willingness to pay for reducing the risk of mortality caused by air pollution from port activities in port city residents. The research involves data collection through a survey utilizing a double dichotomous choice method. The proposed approach differs from existing ones by introducing a new way of estimating the marginal willingness to pay for each additional unit of risk reduction. These estimates are obtained by approximating the corresponding dependency using an elementary function based on five available coordinate points. These coordinates reflect the marginal willingness to pay for each additional unit of risk reduction, as derived from five different risk reduction scenarios presented in the survey. To more accurately assess the declining willingness to pay, an adjustment for cognitive biases is suggested by incorporating questions that evaluate respondents' competency in working with percentage points. It is assumed that the proposed approach will help mitigate the well-known issue in the literature regarding the dependence of VSL estimates on survey design, particularly the level of risk reduction considered. This, in turn, will reduce the manipulability of such studies and significantly enhance trust in their results.