Forecasting

Western Europe has been strongly affected in the last decades by Saharan dust incursions, causing a high PM10 concentration and red rain. In this study, dust events and the performance of seven neural network prediction models, including convolutional neural networks (CNN) and recurrent neural networks (RNN), have been analyzed in a PM10 concentration series from a monitoring station in Córdoba, southern Spain. The models were also assessed here for recursive multi-step prediction over different forecast periods in three different situations: background concentration, a strong dust event, and an extreme dust event. A very important increase in the number of dust events has been identified in the last few years. Results show that CNN models outperform the other models in terms of accuracy for direct 24 h prediction (RMSE values between 10.00 and 10.20 μg/m³), whereas the recursive prediction is only suitable for background concentration in the short term (for 2–5-day forecasts). The assessment and improvement of prediction models might help the development of early-warning systems for these events. From the authors’ perspective, the evaluation of trained models beyond the direct multi-step predictions allowed to fill a gap in this research field, which few articles have explored in depth. Full article

Accurate solar resource forecasting is important because of the inherent variability associated with solar energy and its significant impact on the cost for energy producers. The traditional method applied in solar irradiance forecasting involves two main phases, related to (1) data selection and (2) model selection, training, and evaluation. In this study, we propose a novel end-to-end methodology for solar irradiance forecasting that starts with the search for the data and all of the preprocessing operations involved in obtaining a quality dataset, continuing by imputing missing data with the BERT (Bidirectional Encoder Representations from Transformers) model, and ending with obtaining and evaluating the predicted values. This novel methodology is based on three phases; namely, Phase_1, related to the acquisition and preparation of the data, Phase_2, related to the proposed imputation with a BERT model, and Phase_3, related to the training and prediction with new models based on deep learning. These phases of the proposed methodology can be applied in a disjointed manner, and were used on two public datasets accessible to the scientific community. Each of the proposed phases proved to be valuable for the workflow, and the application of the novel method delivered increases in performance of up to 3 percentage points (3%) when compared to the traditional approach. Full article

This paper deals with a test on forecast bias in predicting independent multinomial outcomes where the predictions are probabilities. The new Likelihood Ratio (and Wald) test extends the familiar Mincer Zarnowitz regression to a multinomial logit model instead of a linear regression. The test is evaluated using various simulation experiments, which indicate that the size and power properties are good, even for small sample sizes, in the sense that the size is close to the used 5% level, and the power quickly reaches 1. We implement the test in an empirical setting on brand choice by individual households. Full article

The methodological framework introduced in this paper, MECOVMA, is a novel framework that guides the application of Machine Learning specifically for marketing predictions within volatile macroeconomic environments. MECOVMA has been developed in response to the identified gaps displayed by existing frameworks—when it comes to consolidation, relevance, interdisciplinarity, and individuality—and in light of the polycrises occurring in the current decade. The methodology to develop MECOVMA comprises three phases: firstly, synthesizing existing frameworks based on their thematic relevance to select MECOVMA’s process steps; secondly, integrating the evidence provided by a systematic literature review to design the content of these process steps; and thirdly, using an expert evaluation, structured through a qualitative content analysis, to validate MECOVMA’s applicability. This leads to the final framework with four overarching P_MECOVMA process steps, guiding the Machine Learning application process in this context with specific tasks. These include, for example, the processing of multidimensional data inputs, complexity reduction in a dynamic environment, and training methods adapted to particular macro-conditions. In addition, features are provided on how Machine Learning can be put into marketing practice, incorporating both narrower statistical- and broader business-oriented evaluations, and iterative feedback loops to mitigate limitations. Full article

This study presents a novel methodology for multi-step Bitcoin (BTC) price prediction by combining advanced stacking-based architectures with temporal attention mechanisms. The proposed Temporal Attention-Enhanced Stacking Network (TAESN) integrates the complementary strengths of diverse machine learning algorithms while emphasizing critical temporal features, leading to substantial improvements in forecasting accuracy over traditional methods. Comprehensive experimentation and robust evaluation validate the superior performance of TAESN across various BTC prediction horizons. Additionally, the model not only demonstrates enhanced predictive accuracy but also offers interpretable insights into the temporal dynamics underlying cryptocurrency markets, contributing to both practical forecasting applications and theoretical understanding of market behavior. Full article

The fluctuations in solar irradiance and temperature throughout the year require an accurate methodology for forecasting the generated current of a PV system based on its specifications. The optimal technique must effectively manage rapid weather fluctuations while maintaining high accuracy in forecasting the performance of a PV panel. This work presents a comparative examination of supervised learning algorithms optimized with particle swarm optimization for estimating photovoltaic output current. The empirical formula’s measured currents are compared with outputs from various neural networks techniques, including feedforward neural networks (FFNNs), the general regression network known as GRNN, cascade forward neural networks also known as CFNNs, and adaptive fuzzy inference systems known as ANFISs, all optimized for enhanced accuracy using the particle swarm optimization (PSO) method. The ground data utilized for these models comprises hourly irradiations and temperatures from 2023, sourced from several places in South Africa. The accuracy levels indicated by statistical error margins from the root mean square error (RMSE), mean bias error (MBE), and mean absolute percentage error (MAPE) imply a universal enhancement in the algorithms’ precision upon optimization. Full article

The growing sports betting market works on the premise that sports are unpredictable, making it more likely to be wrong than right, as the user has to choose between win, draw, or lose. So could football, the world’s most popular sport, be predictable? This article studies this question using deep neural networks to predict the outcome of football matches using publicly available data. Data from 24,760 matches from 13 leagues over 2 to 10 years were used as input for the neural network and to generate a state-of-the-art validated feature, the pi-rating, and the parameters proposed in this work, such as relative attack, defence, and mid power. The data were pre-processed to improve the network’s interpretation and deal with missing or inconsistent data. With the validated pi-rating, data organisation methods were evaluated to find the most fitting option for this prediction system. The final network has four layers with 100, 80, 5, and 3 neurons, respectively, applying the dropout technique to reduce overfitting errors. The results showed that the most influential features are the proposed relative defending, playmaking, and midfield power, and the home team goal expectancy features, surpassing the pi-rating. Finally, the proposed model obtained an accuracy of 52.8% in 2589 matches, reaching 80.3% in specific situations. These results prove that football can be predictable and that some leagues are more predictable than others. Full article

Rainfall forecasting, especially extreme rainfall forecasting, is one of crucial tasks in weather forecasting since it has direct impact on accompanying devastating events such as flash floods and fast-moving landslides. However, obtaining rainfall forecasts with high accuracy, especially for extreme rainfall occurrences, is a challenging task. This study focuses on developing a forecasting model which is capable of forecasting rainfall, including extreme rainfall values. The rainfall forecasting was achieved through sequence learning capability of the Long Short-Term Memory (LSTM) method. The identification of the optimal set of features for the LSTM model was conducted using Random Forest and Granger Causality tests. Then, that best set of features was fed into Stacked LSTM, Bidirectional LSTM, and Encoder-Decoder LSTM models to obtain three days-ahead forecasts of rainfall with the input of the past fourteen days-values of selected features. Out of the three models, the best model was taken through post hoc residual analysis and extra validation approaches. This entire approach was illustrated utilizing rainfall and weather-related measurements obtained from the gauging station located in the city of Ratnapura, Sri Lanka. Originally, twenty-three features were collected including relative humidity, ssunshine hours, and mean sea level pressure. The performances of the three models were compared using $\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}$. The Bidirectional LSTM model outperformed the other methods (RMSE < 5 mm and MAE < 3 mm) and this model has the capability to forecast extreme rainfall values with high accuracy. Full article

The Sacramento–San Joaquin Delta (Delta) is a critical hub of California’s statewide water distribution system. Located at the confluence of California’s two largest rivers, the Sacramento River and the San Joaquin River, the Delta features a complex network of braided channels and over a hundred islands, most of which are located below sea level. The Delta’s complex nature and low-lying topography make it a unique hydrological area pertinent to climate change studies. This paper aims to estimate and explore the potential effects of climate change on the hydrological features of the Delta, especially Net Channel Depletion (NCD), which is one of the main contributors to the Net Delta Outflow (NDO). Downscaled CMIP6 General Circulation Model outputs are used to generate plausible future climate data. The Delta Channel Depletion model (DCD) is used to simulate daily hydrological processes for 61 plausible future climate scenarios. Simulation models are applied to the historical period (1930–2014) and projected future periods (2016–2100). A thorough water balance is computed in the DCD simulation model, offering insights into various elements in the hydrological cycle. Key hydrological features such as crop evapotranspiration, seepage, drainage, and runoff are simulated. Potential changes in NCD, calculated as the sum of diversions and seepage minus drainage, are also examined. The study identified a wide range of increases in NCD across all scenarios in the future period relative to the average of the historical period. These increases are projected to vary from 0.3% up to 20%. Moreover, a spatial analysis conducted across diverse regions of the Delta highlights notable variations in depletion across these areas. The results of this research indicate an anticipated increased stress on water resources, necessitating the adoption of innovative strategies to manage extreme events effectively and ensure the sustainability and resilience of water resource management. Full article

This study presents an innovative machine learning (ML) approach to predicting raw material yield in the leather tanning industry, addressing a critical challenge in production efficiency. Conducted at a tannery in southern Brazil, the research leverages historical production data to develop a predictive model. The methodology encompasses four key stages: data collection, processing, prediction, and evaluation. After rigorous analysis and refinement, the dataset was reduced from 16,046 to 555 high-quality records. Eight ML models were implemented and evaluated using Orange Data Mining software, version 3.38.0, including advanced algorithms such as Random Forest, Gradient Boosting, and neural networks. Model performance was assessed through cross-validation and comprehensive metrics, including Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and Coefficient of Determination (R²). The AdaBoost algorithm emerged as the most accurate predictor, achieving impressive results with an MAE of 0.042, MSE of 0.003, RMSE of 0.057, and R² of 0.331. This research demonstrates the significant potential of ML techniques in enhancing raw material yield forecasting within the tanning industry. The findings contribute to more efficient forecasting processes, aligning with Industry 4.0 principles and paving the way for data-driven decision-making in manufacturing. Full article

This study explores the application of artificial intelligence, specifically ChatGPT-4o, in constructing and managing a portfolio of cybersecurity stocks over the period from Q1 2018 to Q1 2024. Leveraging advanced machine learning models, fundamental analysis, sentiment analysis, and optimization techniques, the AI-driven portfolio significantly outperformed leading cybersecurity ETFs, as well as broader market indices such as the Nasdaq 100 (QQQ) and S&P 500 (SPY). The methodology employed included data collection, stock filtering, predictive modeling using Random Forests and Support Vector Machines (SVMs), sentiment analysis through natural language processing (NLP), and portfolio optimization using Mean-Variance Optimization (MVO), with quarterly rebalancing to ensure responsiveness to evolving market conditions. The AI-selected portfolio achieved a total return of 273%, with strong risk-adjusted performance as demonstrated by key metrics such as the Sharpe ratio, highlighting the effectiveness of an AI-based approach in navigating market complexities and generating superior returns. The results of this study indicate that AI-driven portfolio management can uncover investment opportunities that traditional methods may overlook, offering a competitive edge in the cybersecurity sector and promising enhanced predictive accuracy, efficiency, and overall investment success as AI technologies continue to evolve. Full article

Hydroelectric power is one of the most important renewable energy sources in the world. It currently generates more electricity than all other renewable technologies combined and, according to the International Energy Agency, it is expected to remain the world’s largest source of renewable electricity generation into the 2030s. Thus, despite the increasing focus on more recent energy technologies, such as solar and wind power, it will continue to play a critical role in energy transition. The management of hydropower plants and future planning should be ensured through careful planning based on the suitable forecasting of the future of this energy source. Starting from these considerations, in this paper, we examine the evolution of hydropower with a forecasting analysis for a selected group of countries. We analyze the time-series data of hydropower generation from 1965 to 2023 and apply Innovation Diffusion Models, as well as other models such as Prophet and ARIMA, for comparison. The models are evaluated for different geographical regions, namely the North, South, and Central American countries, the European countries, and the Middle East with Asian countries, to determine their effectiveness in predicting trends in hydropower generation. The models’ accuracy is assessed using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). Through this analysis, we find that, on average, the GGM outperforms the Prophet and ARIMA models, and is more accurate than the Bass model. This study underscores the critical role of precise forecasting in energy planning and suggests further research to validate these results and explore other factors influencing the future of hydroelectric generation. Full article

The rising frequency and severity of droughts requires accurate monitoring and forecasting to reduce the impact on water resources and communities. This study aims to investigate drought monitoring and categorization, while enhancing drought forecasting by using three machine learning models—Artificial Neural Network (ANN), Support Vector Machine (SVM), and Random Forest (RF). The models were trained on the study region’s historic precipitation and temperature data (minimum and maximum) from 1960 to 2021. The Standardized Precipitation Index (SPI) and Standardized Precipitation Evapotranspiration Index (SPEI) were computed for a time scale of 3, 6 and 12 months. The monthly precipitation data were used for creating lag scenarios and were used as input features for the models to improve the models’ performance and reduce overfitting. Statistical parameters like the coefficient of determination (R²), Mean Absolute Error (MAE), Root mean square error (RMSE) and Nash–Sutcliffe Efficiency (NSE) were determined to evaluate the model accuracy. For forecasting, the SPEI3, ANN and SVM models show better performance (R² > 0.9) than the RF models when the 3-month lag data were used as input features. For SPEI6 and SPEI12, the 6-month lag and 12-month lag data, respectively, were needed to increase the models’ accuracy. The models exhibited RMSE values of 0.27 for ANN, 0.28 for SVM, and 0.37 for RF for the SPEI3, indicating the superior performance of the former two. The models’ accuracy increases as the lag period increases for SPI forecasting. Overall, the ANN and SVM models outperformed the RF model for forecasting long-term drought. Full article

This paper focuses on analyzing the relationship between the financial performance of companies and their ability to utilize modern business methods. Financial analysis was conducted using the example of the automobile manufacturer Škoda Auto, with the results providing deeper insights into the company’s financial situation. The companies examined in this study were scored and underwent regression and cluster analyses. A questionnaire focusing on the modernity of advertising in selected companies was answered by 276 respondents. Based on the findings, a model for evaluating the modernity and stability of companies was developed, combining various factors including financial indicators and the adoption of modern technologies. The results indicate that there is a relationship between financial performance and the modernization of companies, although this relationship is not always straightforward. In particular, the operating profit and current ratio emerged as important factors influencing modernization. Overall, it can be concluded that the financial performance and modernization of companies are interconnected, but their relationship is complex and requires further investigation. This paper is an important contribution to understanding company modernization and sets the stage for further studies on this issue. Full article

This study introduces a methodology for labor market foresight through alternative futures. It discusses three alternative scenarios for managerial roles, each exploring varying levels of technological advancement and economic growth, to provide insights into the evolving demands for managerial roles. By drafting a diversified skill portfolio, it is argued that employability skills for managers concern providing education in a combination of areas, such as new technologies, trend analysis, and strategic foresight based on the sector in which the firm operates, negotiation skills and human resources management, contemporary sales techniques, entrepreneurship, and personal growth, including time management, creativity, public speaking skills, and foresight skills. Utilizing responses obtained through an online survey administered in Greece during 2024 to managers and employing principal component analysis (PCA), we establish correlations between skill portfolio composition preferences, foresight analysis, and design of diversified skill portfolios. Diversified skill portfolios are a holistic approach to training, reskilling, and upskilling, including an optimum combination of foundational, complex, digital, green, and always case-fit per occupation or sector of economic activity. Consequently, the insights derived from this study offer a microeconomic perspective regarding the optimal combination of skills for managerial occupations and a macroeconomic perspective concerning the formulation of future training policies for human capital development. Full article

This paper aims to demonstrate how machine deep learning techniques lead to relatively accurate forecasts of quarterly corporate income tax payments. Using quarterly data from Compustat for all U.S. publicly traded corporations from 2000 to 2024, I show that neural nets, the tree method, and random forest models provide robust forecasts despite their encompassing COVID-19 pandemic time periods. The results should be of interest to corporate tax planners, stock analysts, and governments. Full article

Using data that cover the annual period from 1258 to 2023, we studied the link between real gold returns and climate risks. We documented a positive contemporaneous link and a negative predictive link. Our findings further show that the predictive link historically gave rise to significant out-of-sample forecasting gains. The positive contemporaneous link is consistent with the view that investors viewed gold as a safe haven in times of elevated climate risks. The negative predictive link, in turn, is consistent with an overshooting scenario in which the real gold price overshot in response to climate risks, only to return subsequently to a lower value. Our findings should provide important implications for investors and policymakers, given that our analysis covered the longest possible data sample involving the gold market, and hence, was independent of any sample selection bias. Full article

Global food security is under significant threat from climate change, population growth, and resource scarcity. This review examines how advanced AI-driven forecasting models, including machine learning (ML), deep learning (DL), and time-series forecasting models like SARIMA/ARIMA, are transforming regional agricultural practices and food supply chains. Through the integration of Internet of Things (IoT), remote sensing, and blockchain technologies, these models facilitate the real-time monitoring of crop growth, resource allocation, and market dynamics, enhancing decision making and sustainability. The study adopts a mixed-methods approach, including systematic literature analysis and regional case studies. Highlights include AI-driven yield forecasting in European hydroponic systems and resource optimization in southeast Asian aquaponics, showcasing localized efficiency gains. Furthermore, AI applications in food processing, such as plasma, ozone and Pulsed Electric Field (PEF) treatments, are shown to improve food preservation and reduce spoilage. Key challenges—such as data quality, model scalability, and prediction accuracy—are discussed, particularly in the context of data-poor environments, limiting broader model applicability. The paper concludes by outlining future directions, emphasizing context-specific AI implementations, the need for public–private collaboration, and policy interventions to enhance scalability and adoption in food security contexts. Full article

Research shows that Google Trend indices can improve tourism-demand forecasts. Given the impact of the recent pandemic, this may prove to be an important predictor of tourism recovery in countries that are still struggling to recover, including South Africa. The purpose of this paper is firstly, to build on previous research that indicates that Google Trends improves tourism-demand forecasting by testing this within the context of tourism recovery. Secondly, this paper extends previous research by not only including Google Trends in time-series forecasting models but also typical tourism-demand covariates in an econometric specification. Finally, we test the performance of Google Trends in forecasting over a longer time period, because the destination country is a long-haul destination where more lead time may be required in decision-making. Additionally, this research contributes to the body of knowledge by including lower frequency data (quarterly) instead of the higher frequency data commonly used in current research, while also focusing on an important destination country in Africa. Due to the differing data frequencies, the MIDAS modelling approach is used. The MIDAS models are compared to typical time-series and naïve benchmarks. The findings show that monthly Google Trends improve forecasts on lower frequency data. Furthermore, forecasts that include Google Trends are more effective in forecasting one to two quarters ahead, pre-COVID. This trend changed after COVID, when Google Trends led to improved recovery forecasts even over a longer term. Full article