Search Results (36)

In today’s increasingly interconnected digital world, cyber threats have grown in frequency and sophistication, making intrusion detection systems a critical component of modern cybersecurity frameworks. Traditional IDS methods, often based on static signatures and rule-based systems, are no longer sufficient to detect and respond to complex and evolving attacks. To address these challenges, Artificial Intelligence and machine learning have emerged as powerful tools for enhancing the accuracy, adaptability, and automation of IDS solutions. This study presents a novel, hybrid ensemble learning-based intrusion detection framework that integrates deep learning and traditional ML algorithms with explainable artificial intelligence for real-time cybersecurity applications. The proposed model combines an Artificial Neural Network and Support Vector Machine as base classifiers and employs a Random Forest as a meta-classifier to fuse predictions, improving detection performance. Recursive Feature Elimination is utilized for optimal feature selection, while SHapley Additive exPlanations (SHAP) provide both global and local interpretability of the model’s decisions. The framework is deployed using a Flask-based web interface in the Amazon Elastic Compute Cloud environment, capturing live network traffic and offering sub-second inference with visual alerts. Experimental evaluations using the NSL-KDD dataset demonstrate that the ensemble model outperforms individual classifiers, achieving a high accuracy of 99.40%, along with excellent precision, recall, and F1-score metrics. This research not only enhances detection capabilities but also bridges the trust gap in AI-powered security systems through transparency. The solution shows strong potential for application in critical domains such as finance, healthcare, industrial IoT, and government networks, where real-time and interpretable threat detection is vital. Full article

This study examines the relationship between banking sector development (credit and deposits) and economic growth in Ecuador, using quarterly data for the period 2000–2022. An ARDL approach with Bound Test cointegration is employed, incorporating structural breaks using the Bai–Perron test and controlling for macroeconomic shocks. In addition, time transformation methodologies are applied to harmonize the frequency of the series: the monthlyization of GDP is performed using the Chow-Lin method, and the imputation of missing unemployment data using the Kalman filter. The results reveal a significant long-run elasticity between bank deposits and GDP (0.45%), while credits do not present a statistically significant effect, possibly due to high delinquency and institutional weakness. Granger causality tests confirm a unidirectional relationship between banking variables to economic growth. These findings highlight the importance of strengthening financial supervision and improving institutional quality to enhance the effect of bank intermediation. The study provides robust and contextualized empirical evidence relevant to resource-dependent economies with concentrated financial systems, contributing to the debate on the relationship between finance and growth in developing countries. Full article

Time-series forecasting is a cornerstone of decision making in domains such as finance, energy management, and meteorology, where precise predictions drive both economic and operational efficiency. However, traditional time-domain methods often struggle to capture the intricate symmetries and hierarchical dependencies inherent in complex multivariate time-series data. These methods frequently fail to distinguish between global trends and localized fluctuations, limiting their ability to model the multifaceted temporal dynamics that arise across different time scales. To address these challenges, we propose a novel dual-component framework that explicitly leverages the symmetry between long-term trends and short-term fluctuations. Inspired by the principles of signal decomposition, we partition time-series data into a low-frequency stabilization component and a high-frequency fluctuation component. The stabilization component captures inter-variable relationships and global frequency-domain component dependencies through Fourier-transformed frequency-domain representations, variable-oriented attention mechanisms, and dilated causal convolutions. Meanwhile, the fluctuation component models localized dynamics using a multi-granularity structure and time-step attention mechanisms to enhance the sensitivity and robustness to transient variations. By integrating these complementary perspectives, our approach provides a more holistic representation of time-series dynamics. Comprehensive experiments on benchmark datasets from electricity, transportation, and weather domains demonstrate that our method consistently outperforms state-of-the-art models, achieving superior accuracy. Beyond predictive performance, our framework offers a deeper interpretability of temporal behaviors, highlighting its potential for practical applications in complex systems. This work underscores the importance of symmetry-aware modeling in advancing time-series forecasting methodologies. Full article

Applications of high-frequency data, including energy management, economics, and finance, frequently require time-series forecasting characterized by complex seasonality. Recognizing prevailing seasonal trends continues to be difficult, given that the majority of solutions depend on basic decomposition techniques. This study introduces a new approach employing periodograms from spectral density analysis to identify predominant seasonal periods. When analyzing hourly electricity consumption data from Brazil, we identified three significant seasonal patterns: sub-daily (6 h), half-daily (12 h), and daily (24 h). We assessed the predictive efficacy of the BATS, TBATS, and STL + ETS models using these seasonal periods. We performed data analysis and model fitting in R 4.4.1 and used accuracy metrics like MAE, MAPE, and others to compare the models. The STL + ETS model exhibited an enhanced performance, surpassing both BATS and TBATS in energy forecasting. These findings improve our understanding of multiple seasonal patterns, assist us in selecting dominating periods, provide new practical forecasting approaches for time-series analysis, and inform professionals seeking superior forecasting solutions in various fields. Full article

Within the framework of high-frequency volatility modeling, this study investigates the realized volatility spillover dynamics across major cryptocurrencies over an extended period of time. Using a Time-Varying Parameter Vector Autoregression (TVP-VAR) model of the realized volatility (RV), this work constructs the Total Connectedness Index (TCI) and Pairwise Connectedness Index (PCI) to measure the intensity and direction of realized volatility transmission within this digital asset network. Our findings reveal a consistently high level of spillovers among these leading cryptocurrencies, with notable peaks during periods of global market turbulence. Notably, Ethereum emerges as the most influential volatility transmitter, challenging the traditional view of Bitcoin as a primary driver of volatility spillovers. This reflects Ethereum’s pivotal role in decentralized finance (DeFi), decentralized applications (dApps), and its growing trading activity, suggesting a shifting influence in the increasingly diversified cryptocurrency ecosystem. Full article

Industrial linkages play a crucial role in sustaining industrial agglomerations, driving economic growth, and shaping the spatial architecture of economic systems. This study explores the complexity of causal networks within the industrial ecosystems of China and the United States, using high-frequency economic data to compare the interdependencies and causal structures across key sectors. By employing the partial cross mapping (PCM) technique, we capture the dynamic interactions and intricate linkages among industries, providing a detailed analysis of inter-industry causality. Utilizing data from 32 Chinese industries and 11 United States industries spanning 2015 to 2023, our findings reveal that the United States, as a global leader in technology and finance, exhibits a diversified and service-oriented industrial structure, where financial and technology sectors are pivotal to economic propagation. In contrast, China’s industrial network shows higher centrality in heavy industries and manufacturing sectors, underscoring its focus on industrial output and export-led growth. A comparative analysis of the network topology and resilience highlights that China’s industrial structure enhances network stability and interconnectivity, fostering robust inter-industry linkages, whereas the limited nodal points in the United States network constrain its resilience. These insights into causal network complexity offer a comprehensive perspective on the structural dynamics and resilience of the economic systems in both countries. Full article

Supply chain finance (SCF) operations require extensive activities and a high level of information transparency, making them vulnerable to operational issues that pose significant risks of financial loss for commercial banks. Accurately assessing operational risks is crucial for ensuring market stability. This research aims to provide a reliable operational risk assessment tool for commercial banks’ SCF businesses and to deeply examine the features of operational risk events. To achieve these goals, the study explores the dependency structure of risk cells and proposes a quantitative measurement framework for operational risk in SCF. The loss distribution analysis (LDA) is improved to align with the marginal loss distribution of segmented operational risks at both high and low frequencies. A tailored copula function is developed to capture the dependency structure between various risk cells, and the Monte Carlo algorithm is utilized to compute operational risk values. An empirical investigation is conducted using SCF loss data from commercial banks, creating a comprehensive database documenting over 400 entries of SCF loss events from 2012 to 2022. This database is analyzed to identify behaviors, trends, frequencies, and the severity of loss events. The results indicate that fraud risk and compliance risk are the primary sources of operational risks in SCF. The proposed approach is validated through backtesting, revealing a value at risk of CNY 179.3 million and an expected shortfall of CNY 204.9 million at the 99.9% significance level. This study pioneers the measurement of SCF operational risk, offering a comprehensive view of operational risks in SCF and providing an effective risk management tool for financial institutions and policymakers. Full article

The increasing complexity of sustainable development amid financial market regulations has increased the importance of high-quality datasets. However, there is a lack of an integrated approach combining green-finance metrics with the requisite data optimization. This study presents an integrated approach combining green-finance metrics with data optimization. The study uses factorial design methodologies on a sample of 30 firms listed on the Saudi Stock Exchange. Data over five years (2018–2022) were analyzed, focusing on key financial metrics, ESG (environmental, social, and governmental) scores, and sustainability factors. Data analysis used machine-learning models including random forest and XGBoost, Principal Component Analysis (PCA), and regression techniques to evaluate prediction accuracy. The findings revealed that extending the data history from 1–2 to 3–5 years reduced the mean squared error (MSE) by up to 40%, with the XGBoost model achieving an MSE of 0.03 and demonstrating better generalization. In contrast, random forest showed a near-perfect fit with an MSE of 0.00 but risked overfitting. The sampling frequency also affected the accuracy, with weekly and monthly sampling outperforming daily intervals, resulting in an MSE improvement of 15–20%. This study provides a framework for integrating ESG metrics into economic models, aiding policymakers and industry leaders in making informed decisions. The promising results of this study also open avenues for future research and development in sustainable finance and data analysis, offering hope for further progress and innovation. Full article

In the broader landscape of cryptocurrency risk management, this study delves into the nuanced estimation of Value-at-Risk (VaR) for a uniformly weighted portfolio of cryptocurrencies, employing the bivariate Normal Inverse Gaussian distribution renowned for its semi-heavy tails. Utilizing high-frequency data spanning between 1 January 2017 and 25 October 2022, with a primary focus on Bitcoin and Ethereum, our research seeks to accentuate the resilience of VaR methodology as a paramount risk assessment tool. The essence of our investigation lies in advancing the comprehension of VaR accuracy by quantitatively comparing the observed returns of both cryptocurrencies with their corresponding estimated values, with a central theme being the endorsement of the Normal Inverse Gaussian distribution as a potent model for risk measurement, particularly in the domain of high-frequency data. To bolster the statistical reliability of our results, we adopt a forward test methodology, showcasing not only a contribution to the evolution of risk assessment techniques in Finance but also underscoring the practicality of sophisticated distributional models in econometrics. Our findings not only contribute to the refinement of risk assessment methods but also highlight the applicability of such models in precisely modeling and forecasting financial risk within the dynamic realm of cryptocurrencies, epitomized by the case study of Bitcoin and Ethereum. Full article

The object of the investigation is a model of the incomplete financial market. It includes a bank deposit with a known interest rate and basic risky securities. The instant interest rate and volatility are governed by a hidden market regime, represented by some finite-state Markov jump process. The paper presents a solution to two problems. The first one consists of the characterization of the derivatives based on the existing market securities, which are valid to complete the considered market. It is determined that for the market completion, it is sufficient to add the number of derivatives equal to the number of possible market regimes. A generalization of the classic Black–Scholes equation, describing the evolution of the fair derivative price, is obtained along with the structure of a self-financing portfolio, replicating an arbitrary contingent claim in the market. The second problem consists of the online estimation of the market regime, given the observations of both the underlying and derivative prices. The available observations are either a combination of the time-discretized risky security prices or some high-frequency multivariate point processes associated with these prices. The paper presents the numerical algorithms of the market regime tracking for both observation types. The comparative numerical experiments illustrate the high quality of the proposed estimates. Full article

This study bridges finance and physics by applying thermodynamic concepts to model the limit order book (LOB) with high-frequency trading data on the Bitcoin spot. We derive the measures of Market Temperature and Market Entropy from the kinetic and potential energies in the LOB to provide a deeper understanding of order activities and market participant behavior. Market Temperature emerges as a robust indicator of market liquidity, correlating with liquidity measures such as Active Quote Volume, bid–ask spread and match volume. Market Entropy, on the other hand, quantifies the degree of disorder or randomness in the LOB, providing insights into the instantaneous volatility of price in the high-frequency trading market. Our empirical findings not only broaden the theoretical framework of econophysics but also enhance comprehensive understanding of the market microstructure and order book dynamics. Full article

Over the past few years, we have seen an increased need to analyze the dynamically changing behaviors of economic and financial time series. These needs have led to significant demand for methods that denoise non-stationary time series across time and for specific investment horizons (scales) and localized windows (blocks) of time. Wavelets have long been known to decompose non-stationary time series into their different components or scale pieces. Recent methods satisfying this demand first decompose the non-stationary time series using wavelet techniques and then apply a thresholding method to separate and capture the signal and noise components of the series. Traditionally, wavelet thresholding methods rely on the discrete wavelet transform (DWT), which is a static thresholding technique that may not capture the time series of the estimated variance in the additive noise process. We introduce a novel continuous wavelet transform (CWT) dynamically optimized multivariate thresholding method ($Wave{L}_{2}E$). Applying this method, we are simultaneously able to separate and capture the signal and noise components while estimating the dynamic noise variance. Our method shows improved results when compared to well-known methods, especially for high-frequency signal-rich time series, typically observed in finance. Full article

Crude oil has dual attributes of finance and energy. Its price fluctuation significantly impacts global economic development and financial market stability. Therefore, it is necessary to predict crude oil futures prices. In this paper, a hybrid forecast model of EEMD-CNN-ILSTM for crude oil futures price is proposed, which is based on Ensemble Empirical Mode Decomposition (EEMD), Convolutional Neural Network (CNN), and Improved Long Short-Term Memory (ILSTM). ILSTM improves the output gate of Long Short-Term Memory (LSTM) and adds important hidden state information based on the original output. In addition, ILSTM adds the learning of cell state at the previous time in the forget gate and input gate, which makes the model learn more fully from historical data. EEMD decomposes time series data into a residual sequence and multiple Intrinsic Mode Functions (IMF). Then, the IMF components are reconstructed into three sub-sequences of high-frequency, middle-frequency, and low-frequency, which are convenient for CNN to extract the input data’s features effectively. The forecast accuracy of ILSTM is improved efficiently by learning historical data. This paper uses the daily crude oil futures data of the Shanghai Energy Exchange in China as the experimental data set. The EEMD-CNN-ILSTM is compared with seven prediction models: Support Vector Regression (SVR), Multi-Layer Perceptron (MLP), LSTM, ILSTM, CNN-LSTM, CNN-ILSTM, and EEMD-CNN-LSTM. The results of the experiment show the model is more effective and accurate. Full article

Forecasting exchange rates is a complex problem that has benefitted from recent advances and research in machine learning. The main goal of this study is to design and implement a method to improve the learning performance of artificial neural networks with large volumes of data using population-based metaheuristics. The micro-genetic training algorithm is thoroughly analyzed using profiling tools to find bottlenecks. We compare the use of a micro-genetic algorithm to predict changes in currency exchange rates on a data set containing more than 500,000 values. To find the best parameters of neural networks, we propose an improved micro-genetic training algorithm by dividing the training data into mini batches. In this case, the improved micro-genetic algorithm proved to be much faster compared to the standard genetic algorithm, while achieving the same prediction accuracy. This allows for the use of this algorithm for just-in-time predictions of high frequency data. Here, neural network models are first created and validated on an existing data set. Then, the new data values can be added to neural network models and retrained in a short time. Full article

Hardware Trojans (HTs) pose a security threat to the Internet of Things (IoT). Attackers can take control of devices in IoT through HTs, which seriously jeopardize the security of many systems in transportation, finance, healthcare, etc. Since subtle differences in the circuit are reflected in far-field signals emitted by the system, the detection of HT status can be performed by monitoring the radio frequency fingerprinting (RFF) of the transmitting signals. For the detection of HTs, a non-destructive detection method based on RFF is proposed in this paper. Based on the proposed method, the detection of HTs can be achieved without integrating additional devices in the receiver, which reduces associated costs and energy consumption. QPSK and triangular-wave signals are measured and identified via experimentation, and the results validate the proposed method. For identifying the presence and operating state of Trojan, the average accuracy achieved measures as high as 98.7%. Notably, with regard to capturing the moment of Trojan activation in the AES encryption circuit, the accuracy of the proposed method is 100% and can provide warning of the threat in a timely manner. Full article