Search Results (66)

This paper examines the relationship between dependence and independence alternatives in general stochastic processes and explores the duality between the true (yet unknown) stochastic process and the functional representation that fits the observed data. We demonstrate that the solution depends on its historic realizations, challenging existing theoretical frameworks that assume independence between the solution and the history of the true process. Under orthogonality conditions, we investigate parameter spaces within data-generating processes and establish conditions under which data exhibit mean-reverting, random, cyclical, history-dependent, or explosive behaviors. We validate our theoretical framework through empirical analysis of an extensive dataset comprising daily prices from the S&P500, 10-year US Treasury bonds, the EUR/USD exchange rate, Brent oil, and Bitcoin from 1 January 2002 to 1 February 2024. Our out-of-sample predictions, covering the period from 17 February 2019 to 1 February 2024, demonstrate the model’s exceptional forecasting capability, yielding correct predictions with between 73% and 92% accuracy, significantly outperforming naïve and moving average models, which only achieved 47% to 54% accuracy. Full article

Wind energy has emerged as a cornerstone in global efforts to transition to renewable energy, driven by its low environmental impact and significant generation potential. However, the inherent intermittency of wind, influenced by complex and dynamic atmospheric patterns, poses significant challenges for accurate wind speed prediction. Existing approaches, including statistical methods, machine learning, and deep learning, often struggle with limitations such as non-linearity, non-stationarity, computational demands, and the requirement for extensive, high-quality datasets. In response to these challenges, we propose a novel neighborhood preserving cross-dataset data augmentation framework for high-horizon wind speed prediction. The proposed method addresses data variability and dynamic behaviors through three key components: (i) the uniform manifold approximation and projection (UMAP) is employed as a non-linear dimensionality reduction technique to encode local relationships in wind speed time-series data while preserving neighborhood structures, (ii) a localized cross-dataset data augmentation (DA) approach is introduced using UMAP-reduced spaces to enhance data diversity and mitigate variability across datasets, and (iii) recurrent neural networks (RNNs) are trained on the augmented datasets to model temporal dependencies and non-linear patterns effectively. Our framework was evaluated using datasets from diverse geographical locations, including the Argonne Weather Observatory (USA), Chengdu Airport (China), and Beijing Capital International Airport (China). Comparative tests using regression-based measures on RNN, GRU, and LSTM architectures showed that the proposed method was better at improving the accuracy and generalizability of predictions, leading to an average reduction in prediction error. Consequently, our study highlights the potential of integrating advanced dimensionality reduction, data augmentation, and deep learning techniques to address critical challenges in renewable energy forecasting. Full article

Rapid urbanization has accelerated the transformation of community dynamics, highlighting the critical need to understand the interplay between subjective perceptions and objective built environments in shaping life satisfaction for sustainable urban development. Existing studies predominantly focus on linear relationships between isolated factors, neglecting spatial heterogeneity and nonlinear dynamics, which limits the ability to address localized urban challenges. This study addresses these gaps by utilizing multi-scale geographically weighted regression (MGWR) to assess the spatial nonstationarity of subject perceptions and built environment factors while employing gradient-boosting decision trees (GBDT) to capture their nonlinear relationships and incorporating eXtreme Gradient Boosting (XGBoost) to improve predictive accuracy. Using geospatial data (POIs, social media data) and survey responses in Suzhou, China, the findings reveal that (1) proximity to business facilities (β = 0.41) and educational resources (β = 0.32) strongly correlate with satisfaction, while landscape quality shows contradictory effects between central (β = 0.12) and peripheral zones (β = −0.09). (2) XGBoost further quantifies predictive disparities: subjective factors like property service satisfaction (R² = 0.64, MAPE = 3.72) outperform objective metrics (e.g., dining facilities, R² = 0.36), yet objective housing prices demonstrate greater stability (MAPE = 3.11 vs. subjective MAPE = 6.89). (3) Nonlinear thresholds are identified for household income and green space coverage (>15%, saturation effects). These findings expose critical mismatches—residents prioritize localized services over citywide economic metrics, while objective amenities like healthcare accessibility (threshold = 1 km) require spatial recalibration. By bridging spatial nonstationarity (MGWR) and nonlinearity (XGBoost), this study advances a dual-path framework for adaptive urban governance, the community-level prioritization of high-impact subjective factors (e.g., service quality), and data-driven spatial planning informed by nonlinear thresholds (e.g., facility density). The results offer actionable pathways to align smart urban development with socio-spatial equity, emphasizing the need for hyperlocal, perception-sensitive regeneration strategies. Full article

Factors related to catering distribution are typically characterized by local changes, but few studies have quantitatively investigated the inherent spatial nonstationarity correlations. In this study, a multiscale geographically weighted regression (MGWR) model was adopted to locally examine the impact of various factors on catering distribution, which were obtained through a novel method incorporating GeoDetector analysis and exploratory factor analysis (EFA) using point of interest (POI) data. GeoDetector analysis was used to identify the effective variables that truly contribute to catering distribution, and EFA was adopted to extract interpretable latent factors based on the underlying structure of the effective variables and thus eliminate multicollinearity. In our case study in Nanjing, China, four primary factors, namely commuting activities, shopping activities, tourism activities, and gathering activities, were retained from eight categories of POIs with respect to catering distribution. The results suggested that GeoDetector working in tandem with EFA could improve the representativeness of factors and infer POI configuration patterns. The MGWR model explained the most variations (adj. R²: 0.903) with the lowest AICc compared to the OLS regression model and the geographically weighted regression (GWR) model. Mapping MGWR parameter estimates revealed the spatial variability of relationships between various factors and catering distribution. The findings provide useful insights for guiding catering development and optimizing urban functional spaces. Full article

Classification of neurocognitive states from Electroencephalography (EEG) data is complex due to inherent challenges such as noise, non-stationarity, non-linearity, and the high-dimensional and sparse nature of connectivity patterns. Graph-theoretical approaches provide a powerful framework for analysing the latent state dynamics using connectivity measures across spatio-temporal-spectral dimensions. This study applies the graph Koopman embedding kernels (GKKE) method to extract latent neuro-markers of seizures from epileptiform EEG activity. EEG-derived graphs were constructed using correlation and mean phase locking value (mPLV), with adjacency matrices generated via threshold-binarised connectivity. Graph kernels, including Random Walk, Weisfeiler–Lehman (WL), and spectral-decomposition (SD) kernels, were evaluated for latent space feature extraction by approximating Koopman spectral decomposition. The potential of graph Koopman embeddings in identifying latent metastable connectivity structures has been demonstrated with empirical analyses. The robustness of these features was evaluated using classifiers such as Decision Trees, Support Vector Machine (SVM), and Random Forest, on Epilepsy-EEG from the Children’s Hospital Boston’s (CHB)-MIT dataset and cognitive-load-EEG datasets from online repositories. The classification workflow combining mPLV connectivity measure, WL graph Koopman kernel, and Decision Tree (DT) outperformed the alternative combinations, particularly considering the accuracy (91.7%) and F1-score (88.9%), The comparative investigation presented in results section convinces that employing cost-sensitive learning improved the F1-score for the mPLV-WL-DT workflow to 91% compared to 88.9% without cost-sensitive learning. This work advances EEG-based neuro-marker estimation, facilitating reliable assistive tools for prognosis and cognitive training protocols. Full article

Water quality evaluation usually relies on limited state-controlled monitoring data, making it challenging to fully capture variations across an entire basin over time and space. The fine estimation of water quality in a spatial context presents a promising solution to this issue; however, traditional analyses often ignore spatial non-stationarity between variables. To solve the above-mentioned problems in water quality mapping research, we took the Yangtze River as our study subject and attempted to use a geographically weighted random forest regression (GWRFR) model to couple massive station observation data and auxiliary data to carry out a fine estimation of water quality. Specifically, we first utilized state-controlled sections’ water quality monitoring data as input for the GWRFR model to train and map six water quality indicators at a 30 m spatial resolution. We then assessed various geographical and environmental factors contributing to water quality and identified spatial differences. Our results show accurate predictions for all indicators: ammonia nitrogen (NH₃-N) had the lowest accuracy (R² = 0.61, RMSE = 0.13), and total nitrogen (TN) had the highest (R² = 0.74, RMSE = 0.48). The mapping results reveal total nitrogen as the primary pollutant in the Yangtze River basin. Chemical oxygen demand and the permanganate index were mainly influenced by natural factors, while total nitrogen and total phosphorus were impacted by human activities. The spatial distribution of critical influencing factors shows significant clustering. Overall, this study demonstrates the fine spatial distribution of water quality and provides insights into the influencing factors that are crucial for the comprehensive management of water environments. Full article

The development of inner harbors has been accompanied by the destruction of natural landscapes, which in turn has led to numerous ecological problems. However, the temporal and spatial relationships between changes in the inner harbor landscape and ecological effects are not yet clear, and there are relatively few studies at smaller scales such as villages. In this study, we investigated Xieqian Harbor in Xiangshan County, along the eastern coast of China, and then analyzed the landscape change and evolutionary characteristics of the effects of carbon storage, soil conservation, and water yield at the village scale for the years 2000, 2010, and 2020. We then used the geographically and temporally weighted regression (GTWR) model to explore the spatiotemporal relationships between landscape variables and ecological effects. The results showed that the fragmentation and diversity of landscape patches increased from 2000 to 2020 due to reclamation and aquaculture, tourism development, and harbor construction, as reflected by the edge density (ED) and the Shannon diversity index (SHDI), which increased by 11.31% and 2.82%, respectively. This change resulted in a notable reduction of 572.6 thousand tons in carbon sequestration, 853 million tons in soil conservation, and 19 million cubic meters in water yield over the past 20 years. When temporal non-stationarity and spatial heterogeneity were combined, the relationship between landscape change and ecological effects became highly intricate, with varying responses across different time periods and locations. The area-weighted mean patch shape index (AWMSI) was a key factor affecting the three ecological effects. Our research confirmed that there was significant spatiotemporal heterogeneity in the effects of different landscape variables on ecological effects in inner harbors at the village scale. Compared with larger-scale studies, the results of village-scale studies revealed more precisely the impacts of localized landscape changes on ecological effects, providing support for the sustainable management of inner harbors and providing a new approach to integrating GTWR into landscape ecological time–space analysis research. Full article

Traffic crashes significantly contribute to global fatalities, particularly in urban areas, highlighting the need to evaluate the relationship between urban environments and traffic safety. This study extends former spatial modeling frameworks by drawing paths between global models, including spatial lag (SLM), and spatial error (SEM), and local models, including geographically weighted regression (GWR), multi-scale geographically weighted regression (MGWR), and multi-scale geographically weighted regression with spatially lagged dependent variable (MGWRL). Utilizing the proposed framework, this study analyzes severe traffic crashes in relation to urban built environments using various spatial regression models within Leon County, Florida. According to the results, SLM outperforms OLS, SEM, and GWR models. Local models with lagged dependent variables outperform both the global and generic versions of the local models in all performance measures, whereas MGWR and MGWRL outperform GWR and GWRL. Local models performed better than global models, showing spatial non-stationarity; so, the relationship between the dependent and independent variables varies over space. The better performance of models with lagged dependent variables signifies that the spatial distribution of severe crashes is correlated. Finally, the better performance of multi-scale local models than classical local models indicates varying influences of independent variables with different bandwidths. According to the MGWRL model, census block groups close to the urban area with higher population, higher education level, and lower car ownership rates have lower crash rates. On the contrary, motor vehicle percentage for commuting is found to have a negative association with severe crash rate, which suggests the locality of the mentioned associations. Full article

The prevalent utilization of deterministic strategy algorithms in Multi-Agent Deep Reinforcement Learning (MADRL) for collaborative tasks has posed a significant challenge in achieving stable and high-performance cooperative behavior. Addressing the need for the balanced exploration and exploitation of multi-agent ant robots within a partially observable continuous action space, this study introduces a multi-agent centralized strategy gradient algorithm grounded in a local state transition mechanism. In order to solve this challenge, the algorithm learns local state and local state-action representation from local observations and action values, thereby establishing a “local state transition” mechanism autonomously. As the input of the actor network, the automatically extracted local observation representation reduces the input state dimension, enhances the local state features closely related to the local state transition, and promotes the agent to use the local state features that affect the next observation state. To mitigate non-stationarity and reliability assignment issues in multi-agent environments, a centralized critic network evaluates the current joint strategy. The proposed algorithm, NST-FACMAC, is evaluated alongside other multi-agent deterministic strategy algorithms in a continuous control simulation environment using a multi-agent ant robot. The experimental results indicate accelerated convergence and higher average reward values in cooperative multi-agent ant simulation environments. Notably, in four simulated environments named Ant-v2 (2 $\times$ 4), Ant-v2 (2 $\times$ 4d), Ant-v2 (4 $\times$ 2), and Manyant (2 $\times$ 3), the algorithm demonstrates performance improvements of approximately 1.9%, 4.8%, 11.9%, and 36.1%, respectively, compared to the best baseline algorithm. These findings underscore the algorithm’s effectiveness in enhancing the stability of multi-agent ant robot control within dynamic environments. Full article

With the increasing research of deep learning in the EEG field, it becomes more and more important to fully extract the characteristics of EEG signals. Traditional EEG signal classification prediction neither considers the topological structure between the electrodes of the signal collection device nor the data structure of the Euclidean space to accurately reflect the interaction between signals. Graph neural networks can effectively extract features of non-Euclidean spatial data. Therefore, this paper proposes a feature selection method for epilepsy EEG classification based on graph convolutional neural networks (GCNs) and long short-term memory (LSTM) cells. While enriching the input of LSTM, it also makes full use of the information hidden in the EEG signals. In the automatic detection of epileptic seizures based on neural networks, due to the strong non-stationarity and large background noise of the EEG signal, the analysis and processing of the EEG signal has always been a challenging research. Therefore, experiments were conducted using the preprocessed Boston Children’s Hospital epilepsy EEG dataset, and input it into the GCN-LSTM model for deep feature extraction. The GCN network built by the graph convolution layer learns spatial features, then LSTM extracts sequence information, and the final prediction is performed by fully connected and softmax layers. The introduced method has been experimentally proven to be effective in improving the accuracy of epileptic EEG seizure detection. Experimental results show that the average accuracy of binary classification on the CHB-MIT dataset is 99.39%, and the average accuracy of ternary classification is 98.69%. Full article

Sea surface non-stationarity poses significant challenges to sea-surface small target detection, particularly in maintaining a stable false alarm rate (FAR). In dynamic maritime scenarios with non-stationary characteristics, the non-stationarity of sea clutter can easily cause significant changes in the clutter feature space, leading to a notable deviation between the preset FAR and the measured FAR. By analyzing the temporal and spatial variations in sea clutter, we model the relationship between the preset FAR and the measured FAR as a two-parameter linear function. To address the impact of sea surface non-stationarity on FAR, the model parameters are estimated in real time within the environment and used to guide the dynamic adjustment of the decision region. We applied the proposed method to both convex hull and support vector machine (SVM) detectors and conducted experiments using measured X-band sea-detecting datasets. Experiments demonstrate that the proposed method effectively reduces the deviation between the measured mean FAR and the preset FAR. When the preset FAR is 10⁻², the proposed method achieves an average FAR of 1.067 × 10⁻² with the convex hull detector and 1.043 × 10⁻² with the SVM detector. Full article

Many techniques have been proposed to model space-varying observation processes with a nonstationary spatial covariance structure and/or anisotropy, usually on a geostatistical framework. Nevertheless, there is an increasing interest in point process applications, and methodologies that take nonstationarity into account are welcomed. In this sense, this work proposes an extension of a class of spatial Cox process using spatial deformation. The proposed method enables the deformation behavior to be data-driven, through a multivariate latent Gaussian process. Inference leads to intractable posterior distributions that are approximated via MCMC. The convergence of algorithms based on the Metropolis–Hastings steps proved to be slow, and the computational efficiency of the Bayesian updating scheme was improved by adopting Hamiltonian Monte Carlo (HMC) methods. Our proposal was also compared against an alternative anisotropic formulation. Studies based on synthetic data provided empirical evidence of the benefit brought by the adoption of nonstationarity through our anisotropic structure. A real data application was conducted on the spatial spread of the Spodoptera frugiperda pest in a corn-producing agricultural area in southern Brazil. Once again, the proposed method demonstrated its benefit over alternatives. Full article

The accurate prediction of urban road collapses is of paramount importance for public safety and infrastructure management. However, the complex and variable nature of road subsidence mechanisms, coupled with the inherent noise and non-stationarity in the data, poses significant challenges to the development of precise and real-time prediction models. To address these challenges, this paper develops an Adaptive Difference Least Squares Support Vector Regression (AD-LSSVR) model. The AD-LSSVR model employs a difference transformation to process the input and output data, effectively reducing noise and enhancing model stability. This transformation extracts trends and features from the data, leveraging the symmetrical characteristics inherent within it. Additionally, the model parameters were optimized using grid search and cross-validation techniques, which systematically explore the parameter space and evaluate model performance of multiple subsets of data, ensuring both precision and generalizability of the selected parameters. Moreover, a sliding window method was employed to address data sparsity and anomalies, ensuring the robustness and adaptability of the model. The experimental results demonstrate the superior adaptability and precision of the AD-LSSVR model in predicting road collapse timing, highlighting its effectiveness in handling the complex nonlinear data. Full article

The ride-hailing market often experiences significant fluctuations in traffic demand, resulting in supply-demand imbalances. In this regard, the dynamic subsidy strategy is frequently employed by ride-hailing platforms to incentivize drivers to relocate to zones with high demand. However, determining the appropriate amount of subsidy at the appropriate time remains challenging. First, traffic demand exhibits high non-stationarity, characterized by multi-context patterns with time-varying statistical features. Second, high-dimensional state/action spaces contain multiple spatiotemporal dimensions and context patterns. Third, decision-making should satisfy real-time requirements. To address the above challenges, we first construct a Non-Stationary Markov Decision Process (NSMDP) based on the assumption of ride-hailing service systems dynamics. Then, we develop a solution framework for the NSMDP. A change point detection method based on feature-enhanced LSTM within the framework can identify the changepoints and time-varying context patterns of stochastic demand. Moreover, the framework also includes a deterministic policy deep reinforcement learning algorithm to optimize. Finally, through simulated experiments with real-world historical data, we demonstrate the effectiveness of the proposed approach. It performs well in improving the platform’s profits and alleviating supply-demand imbalances under the dynamic subsidy strategy. The results also prove that a scientific dynamic subsidy strategy is particularly effective in the high-demand context pattern with more drastic fluctuations. Additionally, the profitability of dynamic subsidy strategy will increase with the increase of the non-stationary level. Full article

Rainfall serves as a lifeline for crop cultivation in many agriculture-dependent countries including India. Being spatio-temporal data, the forecasting of rainfall becomes a more complex and tedious process. Application of conventional time series models and machine learning techniques will not be a suitable choice as they may not adequately account for the complex spatial and temporal dependencies integrated within the data. This demands some data-driven techniques that can handle the intrinsic patterns such as non-linearity, non-stationarity, and non-normality. Space–Time Autoregressive Moving Average (STARMA) models were highly known for its ability to capture both spatial and temporal dependencies, offering a comprehensive framework for analyzing complex datasets. Spatial Weight Matrix (SWM) developed by the STARMA model helps in integrating the spatial effects of the neighboring sites. The study employed a novel dataset consisting of annual rainfall measurements spanning over 50 (1970–2019) years from 119 different locations (grid of 0.25 × 0.25 degree resolution) of West Bengal, a state of India. These extensive datasets were split into testing and training groups that enable the better understanding of the rainfall patterns at a granular level. The study findings demonstrated a notable improvement in forecasting accuracy by the STARMA model that can exhibit promising implications for agricultural management and planning, particularly in regions vulnerable to climate variability. Full article