Time Series Data Research Articles

The Influence of Natural Resources, Energy Consumption, and Renewable Energy on Economic Growth in ASEAN Region Countries

This research aims to identify and analyze the role of natural resources, CO2 emissions, and renewable energy on the economic growth of countries in the ASEAN region. In this study, we tested whether a dynamic process occurred based on a dynamic panel data model while comparing it with static panel data. The dataset used is ASEAN region countries with time series data from 2000-2021. The analysis tools used are static and dynamic panel data. The research results found that natural resources, FDI, and renewable energy significantly positively affect economic growth in the ASEAN region. Conversely, CO2 has a negative and insignificant effect. These results prove that the ASEAN region's economic development has been clean energy-oriented.

Saudi Arabia’s Green Vision: Examining the Kingdom’s Path to Sustainability, Covering Energy, Economy, Tourism, and Carbon Dynamics

This study examines the Green Vision of Saudi Arabia and its efforts towards achieving sustainability in various domains such as energy, economy, tourism, and carbon emissions. The research uses time-series data from 2002 to 2020 to analyze the dynamic impacts of these variables. To achieve this, it employs a co-integration regression test, Granger causality relationships, and one-way and two-way causal effects between carbon emissions, tourism, economic growth, and energy use. The findings of the study provide valuable insights into the challenges and opportunities for sustainability in the kingdom, considering the goals of SDGs 8 and 13. This research contributes to the ongoing discourse on sustainable development in Saudi Arabia and offers significant insights for lawmakers, entrepreneurs, and researchers to influence the country's future direction. The research aims to guide strategic decision-making in line with Saudi Arabia's commitment to achieving a sustainable and resilient future by examining the interconnections between energy, economic growth, tourism, and carbon emissions.

Comparative Analysis of Bilinear Time Series Models with Time-Varying and Symmetric GARCH Coefficients: Estimation and Simulation

This paper makes a significant contribution by focusing on estimating the coefficients of a sample of non-linear time series, a subject well-established in the statistical literature, using bilinear time series. Specifically, this study delves into a subset of bilinear models where Generalized Autoregressive Conditional Heteroscedastic (GARCH) models serve as the white noise component. The methodology involves applying the Klimko–Nilsen theorem, which plays a crucial role in extracting the asymptotic behavior of the estimators. In this context, the Generalized Autoregressive Conditional Heteroscedastic model of order (1,1) noted that the GARCH (1,1) model is defined as the white noise for the coefficients of the example models. Notably, this GARCH model satisfies the condition of having time-varying coefficients. This study meticulously outlines the essential stationarity conditions required for these models. The estimation of coefficients is accomplished by applying the least squares method. One of the key contributions lies in utilizing the fundamental theorem of Klimko and Nilsen, to prove the asymptotic behavior of the estimators, particularly how they vary with changes in the sample size. This paper illuminates the impact of estimators and their approximations based on varying sample sizes. Extending our study to include the estimation of bilinear models alongside GARCH and GARCH symmetric coefficients adds depth to our analysis and provides valuable insights into modeling financial time series data. Furthermore, this study sheds light on the influence of the GARCH white noise trace on the estimation of model coefficients. The results establish a clear connection between the model characteristics and the nature of the white noise, contributing to a more profound understanding of the relationship between these elements.

Time series diffusion method: A denoising diffusion probabilistic model for vibration signal generation

Diffusion models have demonstrated powerful data generation capabilities in various research fields such as image generation. However, in the field of vibration signal generation, the criteria for evaluating the quality of the generated signal are different from that of image generation and there is a fundamental difference between them. At present, there is no research on the ability of diffusion model to generate vibration signal. In this paper, a Time Series Diffusion Method (TSDM) is proposed for vibration signal generation, leveraging the foundational principles of diffusion models. The TSDM uses an improved U-net architecture with attention block, ResBlock and TimeEmbedding to effectively segment and extract features from one-dimensional time series data. It operates based on forward diffusion and reverse denoising processes for time-series generation. Experimental validation is conducted using single-frequency, multi-frequency datasets, and bearing fault datasets. The results show that TSDM can accurately generate the single-frequency and multi-frequency features in the time series and retain the basic frequency features for the diffusion generation results of the bearing fault series. It is also found that the original DDPM could not generate high quality vibration signals, but the improved U-net in TSDM, which applied the combination of attention block and ResBlock, could effectively improve the quality of vibration signal generation. Finally, TSDM is applied to the small sample fault diagnosis of three public bearing fault datasets, and the results show that the accuracy of small sample fault diagnosis of the three datasets is improved by 32.380%, 18.355% and 9.298% at most, respectively.

Linking carbon storage with land use dynamics in a coastal Ramsar wetland

Coastal wetland ecosystems make an important contribution to the global carbon pool, yet their extent is declining due to aquaculture-related land use changes. We conducted an extensive investigation into the carbon stock and area coverage of macrophytes in a tropical coastal Ramsar wetland, Kolleru in Andhra Pradesh, India. A total of 72 quadrats of size 1 × 1 m2 were laid in the wetland, 19 species of macrophytes were collected and analyzed for carbon content using a CNHS analyzer. To assess changes in the wetland macrophytes, Normalized Difference Vegetation Index (NDVI) was estimated using Landsat time series data from 1975 to 2023. The importance value index (IVI) of macrophytes scored highest for the Ipomoea aquatica (41.4) and the lowest for Ottelia alismoides (1.9). Non-metric multidimensional scaling (NMDS) significantly (r = 0.1905, p = 0.0361) revealed a clear separation of macrophytes in ordination space. ANOVA indicated highly significant (p < 0.0001) variations in the carbon content of aboveground and belowground components of macrophytes. Among the different macrophytes, the highest carbon content was found in Phragmites karka (0.6 g. g−1) and the lowest was recorded in Utricularia stellaris (0.2 g. g−1). On an average, emergents in the Kolleru wetland sequester 1525 ± 181 g C m−2 yr−1, rooted floating species sequester 858 ± 101 g C m−2 yr−1, submerged macrophytes sequester 480 ± 60 g C m−2 yr−1, and free-floating macrophytes sequester 221 ± 90 g C m−2 yr−1. Land cover mapping revealed a decrease in spread of aquatic vegetation from 225.2 km2 in 1975 to 100.6 km2 in 2023. Although macrophytes are vital carbon sinks, the wetland conversion into fishponds has resulted in a loss of 55.3 % of carbon storage. Therefore, immediate restoration of macrophyte cover is vital for the proper functioning of carbon sequestration and mitigation of climate change impacts.

Energy Prices and Their Impact on US Stock Indices: A Wavelet- based Quantile-on-Quantile Regression Approach

This study delves into the effects of crude oil and gas prices on the United States’ (US) conventional, Islamic, and environmental, social, and governance (ESG) stock indices from January 2013 to December 2022. Decomposing original time series data to minimise inherent fluctuations and using the Quantile-on-Quantile (QQ) regression approach presents a nuanced view of how these energy prices impact different stock indices. The findings reveal that crude oil prices have a variable impact on the indices: high prices negatively influence the indices, low prices have a positive effect, and moderate prices yield a moderate positive impact. After data decomposition, this positive influence diminishes in higher quantiles, indicating an emerging neutral effect in stabilised conditions. In contrast, gas prices show a limited impact, with high prices slightly benefiting conventional and ESG indices but less so for the Islamic index. This suggests a more pronounced influence of oil prices on the indices, likely due to the dependence of many listed companies on oil. The study emphasises the importance of considering oil-related risks in investment strategies and highlights the asymmetric impact of crude oil prices on the US stock indices. These findings have significant implications for investors and policymakers. They underscore the need for careful consideration of oil price dynamics in investment decisions and the importance of staying vigilant against shifts in oil prices that could lead to market instability.

Gut-associated functions are favored during microbiome assembly across a major part of C. elegans life.

The microbiome plays a crucial role in host biology. Its functions depend on the microbiome composition that can change during a host's lifetime. To date, the dynamics of microbiome assembly and the resulting functions still need to be better understood. This study introduces a new approach to characterize the functional consequences of microbiome assembly by modeling both the relevance of stochastic processes and metabolic characteristics of microbial community changes. The approach was applied to experimental time-series data obtained for the microbiome of the nematode Caenorhabditis elegans across the major part of its lifetime. Stochastic processes played a minor role, whereas beneficial bacteria as well as gut-associated functions enriched in hosts. This indicates that the host might actively shape the composition of its microbiome. Overall, this study provides a framework for studying microbiome assembly dynamics and yields new insights into C. elegans microbiome functions.

Short-term power load forecasting using SSA-CNN-LSTM method

ABSTRACT The short-term power load forecasting provides an essential foundation for the dispatching management of the power system, which is crucial for enhancing economy and ensuring operational stability. To enhance the precision of the short-term power load forecasting, this paper proposes a hybrid prediction algorithm based on sparrow search algorithm (SSA), convolutional neural network (CNN) and long short-term memory (LSTM). First, feature datasets are constructed based on date information, meteorological data, similar days. The CNN performs effective feature extraction on the data and feeds the results into the LSTM for time series data analysis. Second, eight key parameters are optimized by SSA for improving the prediction precision of the CNN-LSTM prediction model. Simulation results show that the R2 of the proposed model exhibits a substantial enhancement in comparison to other models, reaching 0.9919 and presents a remarkable decrease in MAPE resulting in a value of 1.2%. Furthermore, RMSE and MAE have decreased to 1.17MW and 0.97MW respectively. Therefore, the proposed method has the ability to improve the prediction accuracy, due to the advantages in data mining of CNN, good time series data fitting ability of LSTM, and excellent optimization ability of SSA.

A Deep Learning Approach for Short-Term Electricity Demand Forecasting: Analysis of Thailand Data

Accurate electricity demand forecasting serves as a vital planning tool, enhancing the reliability of management decisions. Apart from that, achieving these aims, particularly in managing peak demand, faces challenges due to the industry’s volatility and the ongoing increase in residential energy use. Our research suggests that employing deep learning algorithms, such as recurrent neural networks (RNN), long short-term memory (LSTM), and gated recurrent units (GRU), holds promise for the accurate forecasting of electrical energy demand in time series data. This paper presents the construction and testing of three deep learning models across three separate scenarios. Scenario 1 involves utilizing data from all-day demand. In Scenario 2, only weekday data are considered. Scenario 3 uses data from non-working days (Saturdays, Sundays, and holidays). The models underwent training and testing across a wide range of alternative hyperparameters to determine the optimal configuration. The proposed model’s validation involved utilizing a dataset comprising half-hourly electrical energy demand data spanning seven years from the Electricity Generating Authority of Thailand (EGAT). In terms of model performance, we determined that the RNN-GRU model performed better when the dataset was substantial, especially in scenarios 1 and 2. On the other hand, the RNN-LSTM model is excellent in Scenario 3. Specifically, the RNN-GRU model achieved an MAE (mean absolute error) of 214.79 MW and an MAPE (mean absolute percentage error) of 2.08% for Scenario 1, and an MAE of 181.63 MW and MAPE of 1.89% for Scenario 2. Conversely, the RNN-LSTM model obtained an MAE of 226.76 MW and an MAPE of 2.13% for Scenario 3. Furthermore, given the expanded dataset in Scenario 3, we can anticipate even higher precision in the results.

An Analytical Study on Growth Performance of Major Food Grain and Oilseed Crops in Context of Food Security: Rajasthan, India

Background: Productivity, as a source of growth has been a significant premise of analytical investigation in economics all along, however it gained an empirical momentum in the recent years only and has become a necessary index to growth. Present paper estimates growth rate of area, production and productivity of food grain and oilseed crops in Rajasthan. Methods: To analyse growth performance of food grain and oilseed crops Rajasthan time series, data on area, production and yield were collected from 1976-77 to 2018-19. The time-series data were collected on area, production and yield of selected crops from the Directorate of Economics and Statistics (DES), Directorate of Agriculture (DOA) and Rajasthan State Agriculture Department. The compound growth rates (CGRs) of area, production and productivity of selected crops in Rajasthan were calculated using the exponential function. Result: The study revealed that production, productivity and area of total food grains was recorded a positive and highly significant annual growth at the rate of 3.18%, 2.19% 0.63% per annum, respectively. The production, productivity and area of total oilseed was observed a positive and highly significant annual growth rate of 7.63%, 2.07% and 4.27% per annum, respectively. The area under barley, gram and sesamum reduced during the study period. Therefore, the technological support through the varietal improvement and extension support is required.

Hydrogeomorphological approach for flood analyses at high- detailed scale: Narrow rivers with broad complex alluvial plains

Fluvial flood risk management requires precise knowledge of flooding processes; thus, there is a need to carry out a detailed hydrological, hydraulic, and geomorphologic analysis to understand floods and the factors that influence them. Although the integration of geomorphological methods and hydrological-hydraulic modelling is known to be crucial for flood risk analysis, it remains a challenge to apply these approaches to the understanding of floods in small and medium-sized catchments, focusing on complex fluvial systems that have been highly modified. In this work, a hydrogeomorphological methodology was developed to analyse fluvial floods in narrow rivers with broad complex alluvial plains. The procedure comprises different phases that combine hydrologic-hydraulic modelling and a geomorphological break-down. Critically, this allows for a detailed analysis of existing relationships among erosional and depositional fluvial landforms and hydraulic parameters, comprising geometric sections, discharges, depths, flood extents and, therefore, flood hazards. The approach involves the analysis of specific flood events beforehand and the study of fluvial floods associated with different probabilities of occurrence afterwards. The information employed in the present study incorporates high- detailed inputs comprising hourly time series of hydrometeorological data, LiDAR data to extract high-precision terrain information, fluvial geomorphic maps and flood event extents based on Sentinel-2 imagery. We demonstrate this methodology in the Carrión River, located in the province of Palencia, Spain, whose characteristics include a wide complex alluvial plain, reaches highly modified by anthropic interventions and recurrent flood episodes. This hydrogeomorphological approach allowed the integration of geomorphological knowledge and hydrologic-hydraulic modelling in flood analysis. As a result, it led to a deeper comprehension of the processes of inundation, which would be challenging to thoroughly understand without incorporating a holistic method, an essential factor for defining effective flood management strategies.

A method for evaluating the risks of glucose dysregulation in daily life by continuous measurement of glucose excursions under reduced glycemic load: a pilot study

Introduction: Continuous glucose monitoring is essential for the management of patients with metabolic syndromes in their daily lives. However, how metabolic risks are accessed using time-series glucose excursion data for healthy individuals remains unclear.Methods: Thus, this study aimed to present a method for evaluating the risks of glucose dysregulation by analyzing glucose responses under reduced glycemic load using a phytochemical alpha-glucosidase inhibitor. A 14-day pilot trial with healthy adult participants was conducted to demonstrate the validity of the method.Results and Discussion: The results showed that ultradian glucose spike packets with lower heights dominantly appeared under reduced glycemic load, replacing postprandial single-lobe glucose spikes with higher heights. Furthermore, the frequency dominance of ultradian spike packets contributed to the reduction of overall mean glucose excursions during the test period under reduced glycemic load. Based on the time-delay model, it was indicated that glycemic load reduction recovers regular glucose–insulin coupling accompanying ultradian oscillations, thereby contributing to the maintenance of glucose homeostasis. The findings of this study indicate that the frequency ratio between pulsatile spike packets and single-lobe spikes could be a marker for evaluating glucose regulation/dysregulation in healthy adults.

Disaggregating National Statistical Data to Assess the Sub-National SDG 6.4.2 Water Stress Indicator by Integrating Satellite Observations and Model Data

Ensuring the sustainable management of water and sanitation for all is the primary goal of Sustainable Development Goal 6 (SDG 6). SDG indicator 6.4.2 (level of water stress) is critical for monitoring the progress toward SDG 6. The assessment of the SDG indicator 6.4.2 is currently based on statistical data at the national scale, i.e., one value for one country, which cannot reflect spatial variability in water stress in a country. The lack of data at sub-national scales limits the assessment of water stress in sub-national regions. This study developed a method of disaggregating national statistical renewable water resources (TRWR) and freshwater withdrawals (TFWW) to estimate the SDG 6.4.2 water stress indicator at a sub-national scale by combining satellite remote sensing data and model simulated data. Remote sensing (RS)-based precipitation (P); the difference between precipitation and evapotranspiration (P-ET); and the difference between precipitation, evapotranspiration, terrestrial water storage change (P-ET-dS), and model-simulated naturized runoff and withdrawal water use were used as spatial and temporal surrogates to disaggregate the national-scale statistics of TRWR and TFWW to the grid scale. Gridded TRWR and TFWW can be used to calculate the water stress of any interest regions. Disaggregated TRWR, TFWW, and water stress estimation were validated at three different spatial scales, from major river basins and provinces to prefectures in China, by comparing the corresponding statistical data. The results show that the disaggregation for TRWR is generally better than for TFWW, and the overall accuracy for water stress estimation can reach up to 91%. The temporal evolution of disaggregated variables also showed good consistency with statistical time series data. The RS-based P-ET and P-ET-dS have great potential for disaggregating TRWR at different spatiotemporal scales, with no obvious differences with the results using the model simulation as a surrogate for the disaggregation of SDG indicator 6.4.2. The disaggregation accuracy can be further improved when the sub-regional statistical data of TRWR and TFWW are applied to the disaggregation approach.

Predicting low-cycle fatigue-induced fracture in reinforcing bars: A CNN-based approach

In low-damage structures such as rocking columns in bridges where the column ends are protected from spalling through confinement, the longitudinal reinforcing bars undergo reversing strain cycles, leading to fracture due to low-cycle fatigue. Several challenges arise due to the unsuitability of strain gauges to measure high strain ranges, in addition to the inability to visually inspect reinforcing bars after seismic events due to the use of confining details at the ends. Building on previous research that identified the fracture of reinforcing bars using column ends’ rotation to estimate reinforcing bars’ strains in conjunction with low-cycle fatigue models, this paper presents substantial development to predict the low-cycle fatigue-induced fracture of reinforcing bars using convolutional neural networks (CNNs) solely from strain time series data. The fracture was identified through strain measurements measured from a shake table testing of a quarter-scale, two-span resilient bridge specimen subjected to seismic excitations at the Earthquake Laboratory at the University of Nevada, Reno. The developed CNN model utilizes the Markov Transition Field technique to encode the strain time series into images and then uses the encoded images as input for channels in the input layer. These images are stacked in sequence to create a 3D array with 11 channels, one for each of the 11 different earthquake excitations that caused the damage. A three-layered CNN architecture with Adam optimizer was employed in training the model, achieving an accuracy of 100 % during training and more than 96 % during testing. To evaluate the model's performance, three distinct training/testing scenarios are proposed. The results demonstrate the efficacy of using CNNs to detect and characterize damage in structural elements using strain data. This approach has the potential to revolutionize the estimation of material fracture due to low-cycle fatigue in scientific fields using only the recoded strains.

MACHINE LEARNING FOR PREDICTING SPREAD OF COVID-19 IN INDONESIA

In previous research, we carried out an analysis using the FBProphet model to predict the COVID-19 outbreak in Indonesia. The application of the FBProphet model to time series data is considered quite good because it produces a MAPE of 22.60% with a linear distribution. Additionally, based on the pattern in the previous dataset and the total number of active cases currently stands at 2,606, in this research we tried to use the Linear Regression (LR) model as a comparison with the FBProphet model by using additional data from the same data source, KAWALCOVID19 website. Data collection started from March 2, 2020 to December 19, 2021. The aim of this research is the same as previous research, namely predicting the spread of COVID-19. The analysis process is carried out by preprocessing the data by validating missing data and validating the format of the data variables. Then carry out descriptive analysis and data visualization so that it can be seen that in this 657 data there is a fluctuates data that non-periodically from July to August 2021. Next, model analysis is carried out using FBProphet and LR and validating the results of each model. The research results are in the form of evaluation metrics where the LR model gets better RMSE, MAE and MAPE values compared to FBProphet, namely 292.91; 178, 81 and 12.79%.

Time and frequency-domain feature fusion network for multivariate time series classification

Multivariate time series classification is a significant research topic in the realm of data mining, which encompasses a wide array of practical applications in domains such as healthcare, energy systems, and traffic. The complex temporal and spatial dependencies inherent in multivariate time series pose challenges to classification tasks. Previous studies have usually focused on the time-domain information of multivariate time series. However, achieving accurate classification using only the time-domain information may be difficult. To address this challenge, a time and frequency-domain feature fusion network (TF-Net) for multivariate time series classification is proposed in this paper. Our model contains two modules, the time-domain module and the frequency-domain module. The time-domain module is used to capture the time-domain features of multivariate time series. It is constructed using CNNs and GCNs, enabling the capture of both temporal and spatial dependencies within the time-domain. The frequency-domain module is used to capture the frequency-domain features of multivariate time series data. In this module, we treat the frequency-domain features as images and innovatively transform the multivariate time series classification task into an image classification task. Our method is able to classify multivariate time series from both the time-domain and frequency-domain, which provides a new perspective for multivariate time series analysis. We conduct extensive experiments on the UAE archive, and the experimental results show that our model achieves the best performance compared to the ten state-of-the-art methods.

Associations between trauma exposure and irritability within the family unit: a network approach.

Pediatric irritability is a pervasive psychiatric symptom, yet its etiology remains elusive. While trauma exposure may contribute to the development of irritability, empirical research is limited. This study examined the prevalence of irritability among trauma-exposed children, identified factors that differentiate trauma-exposed children with and without irritability, and employed a network analysis to uncover associations between irritability and trauma exposure in the family unit. Sample included 676 children (56.3% male, mean age = 9.67 ± 3.7 years) and their parents referred by the Connecticut Department of Children and Families to Fathers for Change - a psychotherapy intervention designed to reduce intimate partner violence (IPV) and child maltreatment. Child's trauma exposure, post-traumatic stress disorder (PTSD) symptoms, and irritability were assessed pre-intervention using self- and caregiver-report. Parents self-reported their childhood and adulthood trauma exposures, PTSD symptoms, irritability, psychopathology, and IPV. Across caregiver- and child-reports, 16%-17% of children exhibited irritability. Irritable children experienced greater trauma exposure, interpersonal violence, emotional abuse, and PTSD severity. They had caregivers, particularly mothers, with greater trauma histories, IPV, and psychopathology. Network analysis revealed 10 nodes directly correlated to child's irritability including child's PTSD severity, parental IPV (specifically psychological violence), and parental psychopathology. Results provide initial empirical evidence that pediatric irritability is linked to trauma exposure, suggesting trauma histories be considered in the diagnosis and treatment of irritability. Interventions addressing caregiver trauma, IPV, and psychopathology may ameliorate pediatric irritability. Future studies could benefit from adopting network approaches with longitudinal or time series data to elucidate causality and points of intervention.

Unsupervised detecting anomalies in multivariate time series by Robust Convolutional LSTM Encoder–Decoder (RCLED)

Monitoring modern industrial systems generates a large amount of multivariate time series data. One of the critical tasks of monitoring these systems is anomaly detection. The auto-encoder has emerged as a promising solution for detecting anomalies in systems lacking explicit anomaly data in their historical records; however, its performance can be sensitive to noisy data. This work aims at developing a robust anomaly detection model based on the autoencoder for multivariate time series. The autoencoder architecture consists of convolution, long short-term memory, and self-attention layers for better extraction of complex features from multivariate time series. In addition, a novel training technique, inherited from the idea of robust principal component analysis, was developed to efficiently train the proposed model when the input data is affected by noise. To evaluate the model’s performance, we tested it on two data sets: synthetic, real-industrial, and three public data. We also compare the performance of our model to that of the other state-of-the-art models. The results show that the proposed model outperforms all the latest models, especially when the noise level is considerable.

Interpretable and explainable hybrid model for daily streamflow prediction based on multi-factor drivers.

Streamflow time series data typically exhibit nonlinear and nonstationary characteristics that complicate precise estimation. Recently, multifactorial machine learning (ML) models have been developed to enhance the performance of streamflow predictions. However, the lack of interpretability within these ML models raises concerns about their inner workings and reliability. This paper introduces an innovative hybrid architecture, the TCN-LSTM-Multihead-Attention model, which combines two layers of temporal convolutional networks (TCN) followed by one layer of long short-term memory (LSTM) units, integrated with a Multihead-Attention mechanism for predicting streamflow with streamflow causation-driven prediction samples (RCDP), employing local and global interpretability studies through Shapley values and partial dependency analysis. The find_peaks method was used to identify peak flow events in the test dataset, validating the model's generality and uncovering the physical causative patterns of streamflow. The results show that (1) compared to the LSTM model with the same hyperparameter settings, the proposed TCN-LSTM-Multihead-Attention hybrid model increased the R2 by 52.9%, 2.5%, 43.1%, and 10.7% respectively at four stations in the test set predictions using RCDP samples. Moreover, comparing the prediction results of the hybrid model under different samples in Hengshan station, the R2 for RCDP increased by 5.06% and 1.22% compared to streamflow autoregressive prediction samples (RAP) and meteorological-soil volumetric water content coupled autoregressive prediction samples (MCSAP) respectively. (2) Historical streamflow data from the preceding 3days predominantly influences predictions due to strong autocorrelation, with flow quantity (Q) typically emerging as the most significant feature alongside precipitation (P), surface soil moisture (SSM), and adjacent station flow data. (3) During periods of low and normal flow, historical data remains the most crucial factor; however, during flood periods, the roles of upstream inflow and precipitation become significantly more pronounced. This model facilitates the identification and quantification of various hydrodynamic impacts on flow predictions, including upstream flood propagation, precipitation, and soil moisture conditions. It also elucidates the model's nonlinear relationships and threshold responses, thereby enhancing the interpretability and reliability of streamflow predictions.

Time series clustering based on latent volatility mixture modeling with applications in finance

Modeling financial time series data poses a significant challenge in the realm of time series analysis. The Autoregressive Conditional Heteroskedasticity (ARCH) model stands out as a potent tool for capturing time-varying volatility and heteroskedasticity in financial data. However, conventional ARCH models display sensitivity to departures from normality, leading to the development of extensions employing more flexible distributions. In this context, we propose a robust enhancement to the mixture of ARCH (MoARCH) model by integrating normal mean–variance mixture (NMVM) distributions to model component errors. The stochastic representation of the proposed model allows for a straightforward implementation of an Expectation Conditional Maximization Either (ECME) algorithm for obtaining Maximum Penalized Likelihood estimates (MPL). To thoroughly evaluate the model, we conduct four simulation studies to explore finite-sample properties, assess MPL estimators, scrutinize model robustness, and evaluate the accuracy of our proposal in fitting, clustering, and forecasting. Practical applications further highlight the effectiveness of our methodology, showcasing successful implementations across diverse real datasets.