Multivariate Time Series Approach Research Articles

Accurate and efficient feature classification of urban public open spaces: A deep learning-based multivariate time-series approach

Urban public open spaces (POS) are pivotal in sustainable urban planning, recognized for their positive impacts on the health of residents and environments. However, understanding their physical features in detail via remote sensing remains challenging due to the small and complex area characteristics. Advanced approaches struggle with the requirement for extensive training datasets, which are difficult to obtain and potentially inaccurate in certain settings. Sparse sampling of training data may offer a solution to these challenges, but its inability to consider useful contextual characteristics from objects remains as a significant barrier.To address these challenges, we propose an innovative methodology utilizing the Multivariate Long Short Term Memory–Fully Convolutional Network (MLSTM-FCN). This approach is adept at capturing subtle temporal and spectral variations from sparsely annotated data, making it particularly suited for optical remote sensing in urban POS. Our novel approach has achieved 97.86% accuracy in classifying POS features into 11 classes using only 1,870 annotated sample points and maintained over 90% of accuracy even with 187 samples or 10% of total samples, significantly outperform comparison models. This methodology allows for detailed and efficient monitoring of POS and enhances management strategies with minimal resources and effort. Our research opens new pathways for precise urban landscape study.

Real-time prediction and detection of contacts between vessels and facilities based on AIS: A multivariate time-series classification approach

Contact casualties between vessels and fixed facilities, such as the recent Baltimore bridge collapse, occur frequently in port areas and narrow waters, resulting in significant losses in property and operational efficiency. However, manual reporting of such contacts is typically prone to errors and delays, while automatic prediction and detection of such contacts remain unresolved. To this end, this study proposes a real-time contact prediction and detection method, powered by multivariate time-series classification and Automatic Identification System (AIS) transmissions. In the off-line training procedure, the historical AIS records are processed into time-series sequences containing five dimensions (i.e., distance to ground, heading, speed over ground, course over ground, navigational status). An LSTM-based classifier is built and trained upon those sequences to differentiate patterns between contact casualties (i.e., positive-class) and normal operations (i.e., negative-class). In the on-line detection procedure, the latest transmitted AIS records are fed into the trained classifier, where the contact casualty is predicted, detected and reported in a real-time manner. A dataset encompassing worldwide AIS records of 150 reported contact casualties and 150 normal operations dated from 01/01/2023 to 31/12/2023, is constructed for method training and validation. Upon comparison, the proposed method outperforms state-of-the-art baselines by achieving the highest F1-score (0.9512), accuracy (0.9500), and AUC (0.9557). It timely predicts and detects 28 out of 30 contact casualties, while the remaining two casualties are detected with a slightly delay, demonstrating high feasibility for real-world application.

Multivariate Bayesian Time-Series Model with Multi-temporal Convolution Network for Forecasting Stock Market During COVID-19 Pandemic

The paper proposes a hybrid algorithm for forecasting multiple correlated time-series data, which consists of two main steps. First, it employs a multivariate Bayesian structural time series (MBSTS) approach as a base step. This method allows for the incorporation of potentially high-dimensional regression components, and it utilizes spike and slab priors to identify a parsimonious model. Second, the algorithm includes a post-model fitting diagnostic step where the residuals from the MBSTS step are processed through a multi-input/output temporal convolutional network (M-TCN) with multiple time scale feature learning. This step serves as an alternative to traditional subjective residual-based diagnostic procedures in time-series analysis, with the aim of improving forecasting accuracy. The key advantage of the M-TCN is its ability to capture sequential information efficiently. The M-TCN expands the field of convolution kernel without increasing the number of parameters, thus enhancing the capacity of model to capture complex sequential patterns. The paper presents two applications showcasing the effectiveness of the proposed hybrid algorithm. First, it utilizes pre-lockdown data from eleven Nifty stock sectoral indices to predict stock price movements, including the initial post-lockdown upturn. In the second application, it focuses on stock market data from pharmaceutical companies involved in manufacturing COVID-19 vaccines. In both cases, sentiment data sourced from newspapers and social media serve as the regression component. Through rigorous analysis, the paper demonstrates that the hybrid model outperforms various benchmark models, including LSTM, Bidirectional Encoder Representations from Transformers (BERT)-based LSTM, Deep Transformer Model, and GRU, among others, in terms of forecasting accuracy. This underscores the utility of the hybrid algorithm, particularly in predicting stock market trends during the COVID-19 pandemic period and its associated market dynamics.

A two-stage adversarial Transformer based approach for multivariate industrial time series anomaly detection

A two-stage adversarial Transformer based approach for multivariate industrial time series anomaly detection

Modeling and Forecasting Closing Prices of some Coal Mining Companies in Indonesia by Using the VAR(3)-BEKK GARCH(1,1) Model

Today, coal is the main source of energy in both developed and developing countries. The use of coal fuel for power generation and industry continues to increase. This research will discuss the closing price relationship model for the share prices of two coal companies in Indonesia, namely ABM and IND_E, from January 2018 to July 2023. The modeling used is a multivariate time series approach. From the results of the data analysis, the best model that fits the data is the VAR(3)-BEKK GARCH(1,1). Based on this best model, further analysis of Granger causality, impulse response function (IRF), and forecasting for the next 30 periods as well as the proportion of prediction error covariance are discussed.

A dimension reduction factor approach for multivariate time series with long-memory: a robust alternative method

A dimension reduction factor approach for multivariate time series with long-memory: a robust alternative method

Analysis Forecasting of Gasoline Prices in Some ASEAN Countries by Using State Space Representation on Vector Autoregressive Model

Research on the price of gasoline has become a topic of research that has been carried out by many researchers. The topic is interesting because the price of gasoline has a relationship with many aspects of people's lives. This study aims to examine the relationship pattern of gasoline prices in several ASEAN countries: Indonesia, Malaysia, and Vietnam, and to make gasoline price forecasting in these three countries for the next 12 months. This study uses a multivariate time series approach; first, the best vector autoregressive (VAR(p)) model will be built based on Akaike's Information Criterion (AIC). Based on the best VAR(p) model, granger-causality analysis is discussed, and for forecasting gasoline prices, a state space model will be developed based on the best VAR(p). State vectors are built based on canonical correlation analysis. Based on the results of granger causality analysis, gasoline prices in Indonesia are affected by past gasoline prices in Vietnam; gasoline prices in Malaysia are affected by past gasoline prices in Indonesia and Vietnam. The results of forecasting analysis for the next 12 months using the state space model show that gasoline prices in Indonesia for the next 12 months tend to have a downward trend; gasoline prices in Malaysia for the next 12 months tend to have an upward trend; and the price of gasoline in Vietnam for the next 12 months tends to have an upward trend for the first 6 months and then has a downward trend for the next 6 months.

Emotional states predict cellular immune system activity under conditions of life as it is lived: A multivariate time-series analysis approach.

The relationship between emotional states and immune system activity is characterized by bidirectional influences; however, limited information is available regarding the temporal dynamics of these effects. The goal of this investigation was to examine how these psychoimmunological interdependencies unfold over time under conditions of "life as it is lived". For this purpose, three healthy women collected their entire urine over a period of approximately two months at 12-h intervals (8 am-8 pm, 8 pm-8 am), resulting in a total of 112 to 126 consecutive measurements per subject. In addition, among other regular psychological assessments, the subjects completed the EWL-60-S, an emotional state questionnaire, each morning and evening. To assess the extent of T-helper type 1 immune activation, the neopterin per creatinine concentration was measured in the urine samples using high-pressure liquid chromatography. The dynamic relationships between the time series of the six emotional states (performance-related activity, general inactivity, extraversion/introversion, general feeling of comfort, emotional irritation, anxiety/depressiveness) and urinary neopterin levels were estimated in vector-autoregressive models and evaluated using Granger-causality tests, impulse-response functions and forecast error variance decompositions. The findings showed that emotional states explained up to 20% of the variance of urinary neopterin per creatinine levels, whereby most of the effects occurred within a period of approximately three days. Across all subjects, increases in anxiety/depressiveness and extraversion led to increases in neopterin levels, while a general feeling of comfort led to decreases in neopterin. These results emphasize the importance of the interdependencies between emotional states and immune system activity and showcase the potential that intensive longitudinal study designs offer for psychoneuroimmunology.

Motor imagery EEG signal classification with a multivariate time series approach

BackgroundElectroencephalogram (EEG) signals record electrical activity on the scalp. Measured signals, especially EEG motor imagery signals, are often inconsistent or distorted, which compromises their classification accuracy. Achieving a reliable classification of motor imagery EEG signals opens the door to possibilities such as the assessment of consciousness, brain computer interfaces or diagnostic tools. We seek a method that works with a reduced number of variables, in order to avoid overfitting and to improve interpretability. This work aims to enhance EEG signal classification accuracy by using methods based on time series analysis. Previous work on this line, usually took a univariate approach, thus losing the possibility to take advantage of the correlation information existing within the time series provided by the different electrodes. To overcome this problem, we propose a multivariate approach that can fully capture the relationships among the different time series included in the EEG data. To perform the multivariate time series analysis, we use a multi-resolution analysis approach based on the discrete wavelet transform, together with a stepwise discriminant that selects the most discriminant variables provided by the discrete wavelet transform analysisResultsApplying this methodology to EEG data to differentiate between the motor imagery tasks of moving either hands or feet has yielded very good classification results, achieving in some cases up to 100% of accuracy for this 2-class pre-processed dataset. Besides, the fact that these results were achieved using a reduced number of variables (55 out of 22,176) can shed light on the relevance and impact of those variables.ConclusionsThis work has a potentially large impact, as it enables classification of EEG data based on multivariate time series analysis in an interpretable way with high accuracy. The method allows a model with a reduced number of features, facilitating its interpretability and improving overfitting. Future work will extend the application of this classification method to help in diagnosis procedures for detecting brain pathologies and for its use in brain computer interfaces. In addition, the results presented here suggest that this method could be applied to other fields for the successful analysis of multivariate temporal data.

Inter-regional dependence of J-REIT stock prices: A heteroscedasticity-robust time series approach

This paper investigates the dynamic interdependence between the stock returns of geographically non-overlapping Japanese real estate investment trusts (J-REITs), where the property type and a market return are controlled. We take a multivariate time series approach, allowing for conditional heteroscedasticity of unknown form. We find significant impacts of central J-REITs on local J-REITs in conditional mean, a potential signal of arbitrage opportunities. After the COVID-19 crisis, the central-to-local impacts have become stronger for all property types considered: office, residential, and hotel. This empirical result is consistent with the consensus that portfolio diversification is harder to achieve during a period of turmoil.

Multiway clustering with time-varying parameters

This paper proposes a clustering approach for multivariate time series with time-varying parameters in a multiway framework. Although clustering techniques based on time series distribution characteristics have been extensively studied, methods based on time-varying parameters have only recently been explored and are missing for multivariate time series. This paper fills the gap by proposing a multiway approach for distribution-based clustering of multivariate time series. To show the validity of the proposed clustering procedure, we provide both a simulation study and an application to real air quality time series data.

Predictive multi-watershed flood monitoring using deep learning on integrated physical and social sensors data

This paper presents a deep learning model based on the integration of physical and social sensors data for predictive watershed flood monitoring. The data from flood sensors and 3-1-1 reports data are mapped and fused through a multivariate time series approach. This data format is able to increase the data availability (partially due to sparsely installed physical sensors and fewer reported flood incidents in less urbanized areas) and capture both spatial and temporal interactions between different watersheds and historical events. We use Harris County, TX as the study site and obtained seven historical flood events data for training, validating, and testing the flood prediction model. The model predicts the flood probability of each watershed in the next 24 hours. By comparing the flood prediction performance of three different datasets (i.e., flood sensor only, 3-1-1 reports only, and integrated dataset), we conclude that the physical-social data integrated approach can better predict the flood with an accuracy of 0.825, area under the receiver operating characteristics curve (AURC) of 0.902, area under the precision-recall curve (AUPRC) of 0.883, area under the F-measure curve (AUFC) of 0.762, and Max. F-measure of 0.788.

Variational transformer-based anomaly detection approach for multivariate time series

Due to the strategic importance of satellites, the safety and reliability of satellites have become more important. Sensors that monitor satellites generate lots of multivariate time series, and the abnormal patterns in the multivariate time series may imply malfunctions. The existing anomaly detection methods for multivariate time series have poor effects when processing the data with few dimensions or sparse relationships between sequences. This paper proposes an unsupervised anomaly detection model based on the variational Transformer to solve the above problems. The model uses the Transformer's self-attention mechanism to capture the potential correlations between sequences and capture the multi-scale temporal information through the improved positional encoding and up-sampling algorithm. Then, the model comprehensively considers the extracted features through the residual variational autoencoder to perform effective anomaly detection. Experimental results on a real dataset and two public datasets show that the proposed method is superior to the mainstream and state-of-the-art methods.

An improved monarch butterfly optimization based multivariate fuzzy time series approach for forecasting GDP of India

An improved monarch butterfly optimization based multivariate fuzzy time series approach for forecasting GDP of India

Forecasting the Romanian Unemployment Rate in Time of Health Crisis-A Univariate vs. Multivariate Time Series Approach.

Economic crises cause significant shortages in disposable income and a sharp decline in the living conditions, affecting healthcare sector, hitting the profitability and sustainability of companies leading to raises in unemployment. At micro level, these sharp decreases in earnings associated with unemployment and furthermore with the lack of social protection will impact the quality of life and finally the health of individuals. In time of crisis, it becomes vital to support not only the critical sectors of the economy, the assets, technology, and infrastructure, but to protect jobs and workers. This health crisis has hit hard the jobs dynamics through unemployment and underemployment, the quality of work (through wages, or access to social protection), and through the effects on specific groups, with a higher degree of vulnerability to unfavorable labor market outcomes. In this context, providing forecasts as recent as possible for the unemployment rate, a core indicator of the Romanian labor market that could include the effects of the market shocks it becomes fundamental. Thus, the paper aims to offer valuable forecasts for the Romanian unemployment rate using univariate vs. multivariate time series models for the period 2021–2022, highlighting the main patterns of evolution. Based on the univariate time series models, the paper predict the future values of unemployment rate based on its own past using self-forecasting and implementing ARFIMA and SETAR models using monthly data for the period January 2000–April 2021. From the perspective of multivariate time series models, the paper uses VAR/VECM models, analyzing the temporal interdependencies between variables using quarterly data for the period 2000Q1–2020Q4. The empirical results pointed out that both SETAR and VECM provide very similar results in terms of accuracy replicating very well the pre-pandemic period, 2018Q2–2020Q1, reaching the value of 4.1% at the beginning of 2020, with a decreasing trend reaching the value of 3.9%, respectively, 3.6% at the end of 2022.

Regional Interdependence of the Japan REIT Market: A Heteroscedasticity-Robust Time Series Approach

This paper investigates the dynamic interdependence between the stock returns of regionally disjoint Japanese real estate investment trusts (REITs), where the property type and a market return are controlled. We take a multivariate time series approach with the error term being allowed to have conditional heteroscedasticity of unknown form. We find significant spillover effects from central to local areas in conditional mean, a potential signal of arbitrage opportunities. The spillover effects have become stronger after the COVID-19 crisis for the office and hotel sectors, but not for the residential sector. This contrast suggests that a geographic diversification strategy within residential REIT securities should be more effective than that within office or hotel, especially during a period of turmoil.

Memristor variability and stochastic physical properties modeling from a multivariate time series approach

A powerful time series analysis modeling technique is presented to describe cycle-to-cycle variability in memristors. These devices show variability linked to the inherent stochasticity of device operation and it needs to be accurately modeled to build compact models for circuit simulation and design purposes. A new multivariate approach is proposed for the reset and set voltages that accurately describes the statistical data structure of a resistive switching series. Experimental data were measured from advanced hafnium oxide based devices. The models reproduce the experiments correctly and a comparison of the multivariate and univariate approaches is shown for comparison.

A Divide-and-Conquer–based Early Classification Approach for Multivariate Time Series with Different Sampling Rate Components in IoT

In the era of the Internet of Things (IoT), the sensor-based devices produce the Multivariate Time Series (MTS). A classification approach helps to predict the class label of an incoming MTS. Due to the large dimension and different sampling rate of the sensors in a given MTS, a classifier takes time to predict the class label. Some IoT applications may require early prediction of the class label where the classifier starts the prediction once the minimum number of data points are collected. In this article, we address the problem of early prediction of the class label of an MTS in IoT. This work considers the sensors with different sampling rate to generate the MTS. Each sensor generates a time series (component) of the MTS. We propose a Divide-and-Conquer–based early classification approach for classifying such MTS. The approach constructs an ensemble classifier using a probabilistic classifier and hierarchical clustering. The ensemble classifier employs a Divide-and-Conquer method to handle the different sampling rate components during the prediction of class label. The experimental results show that our approach significantly outperforms the existing approaches on real-world datasets using various evaluation metrics.

Deep Learning Applications with Practical Measured Results in Electronics Industries

This editorial introduces the Special Issue, entitled “Deep Learning Applications with Practical Measured Results in Electronics Industries”, of Electronics. Topics covered in this issue include four main parts: (I) environmental information analyses and predictions, (II) unmanned aerial vehicle (UAV) and object tracking applications, (III) measurement and denoising techniques, and (IV) recommendation systems and education systems. Four papers on environmental information analyses and predictions are as follows: (1) “A Data-Driven Short-Term Forecasting Model for Offshore Wind Speed Prediction Based on Computational Intelligence” by Panapakidis et al.; (2) “Multivariate Temporal Convolutional Network: A Deep Neural Networks Approach for Multivariate Time Series Forecasting” by Wan et al.; (3) “Modeling and Analysis of Adaptive Temperature Compensation for Humidity Sensors” by Xu et al.; (4) “An Image Compression Method for Video Surveillance System in Underground Mines Based on Residual Networks and Discrete Wavelet Transform” by Zhang et al. Three papers on UAV and object tracking applications are as follows: (1) “Trajectory Planning Algorithm of UAV Based on System Positioning Accuracy Constraints” by Zhou et al.; (2) “OTL-Classifier: Towards Imaging Processing for Future Unmanned Overhead Transmission Line Maintenance” by Zhang et al.; (3) “Model Update Strategies about Object Tracking: A State of the Art Review” by Wang et al. Five papers on measurement and denoising techniques are as follows: (1) “Characterization and Correction of the Geometric Errors in Using Confocal Microscope for Extended Topography Measurement. Part I: Models, Algorithms Development and Validation” by Wang et al.; (2) “Characterization and Correction of the Geometric Errors Using a Confocal Microscope for Extended Topography Measurement, Part II: Experimental Study and Uncertainty Evaluation” by Wang et al.; (3) “Deep Transfer HSI Classification Method Based on Information Measure and Optimal Neighborhood Noise Reduction” by Lin et al.; (4) “Quality Assessment of Tire Shearography Images via Ensemble Hybrid Faster Region-Based ConvNets” by Chang et al.; (5) “High-Resolution Image Inpainting Based on Multi-Scale Neural Network” by Sun et al. Two papers on recommendation systems and education systems are as follows: (1) “Deep Learning-Enhanced Framework for Performance Evaluation of a Recommending Interface with Varied Recommendation Position and Intensity Based on Eye-Tracking Equipment Data Processing” by Sulikowski et al. and (2) “Generative Adversarial Network Based Neural Audio Caption Model for Oral Evaluation” by Zhang et al.

Generalized discrete autoregressive moving‐average models

AbstractThis article proposes the generalized discrete autoregressive moving‐average (GDARMA) model as a parsimonious and universally applicable approach for stationary univariate or multivariate time series. The GDARMA model can be applied to any type of quantitative time series. It allows to compute moment properties in a unique way, and it exhibits the autocorrelation structure of the traditional ARMA model. This great flexibility is obtained by using data‐specific variation operators, which is illustrated for the most common types of time series data, such as counts, integers, reals, and compositional data. The practical potential of the GDARMA approach is demonstrated by considering a time series of integers regarding votes for a change of the interest rate, and a time series of compositional data regarding television market shares.