Time Series Models Research Articles

Designing of Airspeed Measurement Method for UAVs Based on MEMS Pressure Sensors.

Airspeed measurement is crucial for UAV control. To achieve accurate airspeed measurements for UAVs, this paper calculates airspeed data by measuring changes in air pressure and temperature. Based on this, a data processing method based on mechanical filtering and the improved AR-SHAKF algorithm is proposed to indirectly measure airspeed with high precision. In particular, a mathematical model for an airspeed measurement system was established, and an installation method for the pressure sensor was designed to measure the total pressure, static pressure, and temperature. Secondly, the measurement principle of the sensor was analyzed, and a metal tube was installed to act as a mechanical filter, particularly in cases where the aircraft has a significant impact on the gas flow field. Furthermore, a time series model was used to establish the sensor state equation and the initial noise values. It also enhanced the Sage-Husa adaptive filter to analyze the unavoidable error impact of initial noise values. By constraining the range of measurement noise, it achieved adaptive noise estimation. To validate the superiority of the proposed method, a low-complexity airspeed measurement device based on MEMS pressure sensors was designed. The results demonstrate that the airspeed measurement device and the designed velocity measurement method can effectively calculate airspeed with high measurement accuracy and strong interference resistance.

Development and Comparison of InSAR-Based Land Subsidence Prediction Models

Land subsidence caused by human engineering activities is a serious problem worldwide. We selected Qian’an County as the study area to explore the evolution of land subsidence and predict its deformation trend. This study utilized synthetic aperture radar interferometry (InSAR) technology to process 64 Sentinel-1 data covering the area, and high-precision and high-resolution surface deformation data from January 2017 to December 2021 were obtained to analyze the deformation characteristics and evolution of land subsidence. Then, land subsidence was predicted using the intelligence neural network theory, machine learning methods, time-series prediction models, dynamic data processing techniques, and engineering geology of ground subsidence. This study developed three time-series prediction models: a support vector regression (SVR), a Holt Exponential Smoothing (Holt) model, and multi-layer perceptron (MLP) models. A time-series prediction analysis was conducted using the surface deformation data of the subsidence funnel area of Zhouzi Village, Qian’an County. In addition, the advantages and disadvantages of the three models were compared and analyzed. The results show that the three developed time-series data prediction models can effectively capture the time-series-related characteristics of surface deformation in the study area. The SVR and Holt models are suitable for analyzing fewer external interference factors and shorter periods, while the MLP model has high accuracy and universality, making it suitable for predicting both short-term and long-term surface deformation. Ultimately, our results are valuable for further research on land subsidence prediction.

Explaining state efforts to create Total Maximum Daily Load (TMDL) agreements

AbstractObjectivesWith the rejuvenated emphasis on nonpoint pollution under the Water Quality Act (WQA) of 1987, Environmental Protection Agency (EPA) began to face an onslaught of lawsuits designed to pressure the EPA to enforce the requirements of Section 319 of the WQA to address nonpoint pollution. Known as Total Maximum Daily Load (TMDL) agreements, the purpose of these plans was to limit the amount of polluted runoff reaching a state's waterways. While some states took a proactive stance on these plans, other states resisted the implementation of Section 319. This article seeks to understand state choices in the development and implementation of TMDL agreements.MethodsUtilizing a data set spanning state‐level data from 2000 to 2020, we test a novel cross‐sectional time series model employing the number agreements entered into by a state as the dependent variable.ResultsWe find that both political and need explanations are generally supported, while policy need explanations are somewhat more promising.ConclusionsTaken together, the models offer several insights into state choices around TMDL creation. The political model is the weakest, suggesting that TMDLs are not overtly political. Policy needs seem to play a more critical role in the preponderance of TMDL agreements than partisan politics.

Forecasting Cost Risks of Corn and Soybean Crops through Monte Carlo Simulation

Considering that investing in the production of corn and soybeans is conditioned by production costs and several risks, the objective of this research work was to develop a simulation model for the prediction of the production costs of these commodities, considering the variability and correlation of key variables. The descriptive analysis of the data focused on measures such as mean, standard deviation, and coefficient of variation. To evaluate the relationship between commodity and input prices, Spearman’s demonstration coefficient and the coefficient of determination (R2) were used. A Monte Carlo simulation (MCS) was used to evaluate the variation in production costs and net revenues. The Predictor tool was used to make predictions based on historical data and time series models. This study was made for the period between 2018 and 2022 based on data provided by fifty companies from the state of São Paulo, Brazil. The results showed that the production cost/ha of corn faces a high-cost risk, particularly when production and market conditions are characterized by high levels of volatility, uncertainty, complexity, and ambiguity. The model proposed forecasts prices more accurately, as it considers the variation in the costs of inputs that most significantly influence the costs of corn and soybean crops.

Assessing the accuracy of directional forecasts

ABSTRACT Directional predictions are used in a variety of settings, including economics, finance, meteorology and medicine. Since the objectives in each setting can be quite different, appropriate tests for assessing the predictions are necessary. After summarizing existing tests, we propose new tests to assess the accuracy for cases where getting extreme events wrong is particularly costly and where there is serial dependence in small samples. We show in simulations that the latter test has a more favourable power-size trade-off relative to existing tests. The new tests are then applied to the directional EUR/USD forecasts in the ifo-Institute’s World Economic Survey and to the point forecasts in the Philadelphia Fed’s Survey of Professional Forecasters. For the former, the new tests show that while there is profitability and predictive accuracy in some cases, a simple time-series model is not beaten. For the latter, extreme values are not the main driver of the predictive value up to two quarters ahead. This shows that the additional tests allow a more precise determination of predictive accuracy and economic value and where it is coming from.

A time series modelling and forecasting of under-5 mortality rate in Nigeria

A time series modelling and forecasting of under-5 mortality rate in Nigeria

Drafting restraint: Are military recruitment policies associated with interstate conflict initiation?

Are countries that use conscription more restrained in their use of military force? A common argument holds that military conscription restrains leaders from using force because it increases the political cost of war and distributes them more evenly and broadly across the population. Despite this intuition, empirical evidence to support it is at best inconclusive. This article introduces a novel perspective on the relationship between military recruitment (MR) policies and conflict initiation (CI) by arguing that the military’s size relative to society – its military participation rate (MPR) – is an important and overlooked part of this story. MPR is a more direct measure of the population’s exposure to the costs of war, but high MPR may also increase CI by enhancing military capacity. By incorporating MPR into the analysis of CI, both independently and in interaction with conscription, this article provides a more comprehensive understanding of how MR practices shape CI. It tests these new hypotheses about the relationship between MPR, conscription and CI using a variety of time-series models that cover all country-years from 1816 to 2011. The findings do not support the conventional wisdom, instead revealing that neither conscription nor volunteerism is independently associated with restrained initiation of military conflicts abroad. On the contrary, these recruitment practices are more likely to be associated with an increase in the likelihood of CI. These findings indicate that we should be skeptical of traditional arguments that assume conscription leads to restraint in the use of force, either independently or conditional on MPR. These counterintuitive results underscore the need for additional research on the complex relationship between MR practices, civil–military relations and foreign policy.

Time-variations of wave energy and forecasting power availability at a site in Fiji using time-series, regression and ANN techniques

ABSTRACT Recently, there has been a shift in the global energy landscape to move to reliable, clean, and eco-friendly renewable energy sources to address global issues such as climate change and greenhouse gas emissions. One such energy source is wave energy; researchers attempt to develop models that can accurately forecast the availability of wave energy as an alternative energy source. In this paper, an Artificial Neural Network (ANN) model along with statistical models such as time series models, and regression models are proposed for forecasting wave energy at a site in Fiji using the wave height and wave period as the independent variables. The performance of the proposed models developed is compared using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and the goodness-of-fit ( R 2 ) value. The proposed model is then further benchmarked with the naïve model. The empirical results reveal that the proposed ANN model outclassed all the other models and was more efficient and accurate in forecasting wave energy than the regression and time series models. By accurate wave modelling and by incorporating impedance matching, maximum power generation can be achieved.

IP3T: an interpretable multimodal time-series model for enhanced gait phase prediction in wearable exoskeletons.

Wearable exoskeletons assist individuals with mobility impairments, enhancing their gait and quality of life. This study presents the iP3T model, designed to optimize gait phase prediction through the fusion of multimodal time-series data. The iP3T model integrates data from stretch sensors, inertial measurement units (IMUs), and surface electromyography (sEMG) to capture comprehensive biomechanical and neuromuscular signals. The model's architecture leverages transformer-based attention mechanisms to prioritize crucial data points. A series of experiments were conducted on a treadmill with five participants to validate the model's performance. The iP3T model consistently outperformed traditional single-modality approaches. In the post-stance phase, the model achieved an RMSE of 1.073 and an R2 of 0.985. The integration of multimodal data enhanced prediction accuracy and reduced metabolic cost during assisted treadmill walking. The study highlights the critical role of each sensor type in providing a holistic understanding of the gait cycle. The attention mechanisms within the iP3T model contribute to its interpretability, allowing for effective optimization of sensor configurations and ultimately improving mobility and quality of life for individuals with gait impairments.

A time-series deep learning model for predicting concrete shrinkage and creep verified with in-situ and laboratory test data

This paper proposed a time series prediction method for concrete shrinkage and creep based on deep learning and the accuracy of the method was verified by field tests. Based on the experiment data in the Northwestern University (NU database), a deep learning model considering the material, environment, and history was constructed by integrating Gated Recurrent Units and Self-Attention mechanism (GSA model). To validated the model, short-term indoor and outdoor field tests of concrete shrinkage and creep were conducted. The proposed GSA model outperformed the LSTM model for both creep and shrinkage predictions. Moreover, the comparison between the GSA model prediction and the outdoor test data indicated that the model can accurately capture the real-time outdoor temperature and humidity variation through historical data. To predict the ten-year shrinkage and creep of concrete, a recursive multi-step method was proposed to extrapolate the short-term predictions of the GSA model. The effectiveness of the recursive multi-step method was validated against the long-term test data in the NU database. Overall, the proposed GSA model demonstrated high accuracy in both short-term and long-term predictions.

Predicting the Path of Insurgency: Data-Driven Strategies to Counter Boko Haram in Nigeria

Background While Boko Haram insurgency’s dangers are well documented, existing research lacks methods for effective monitoring and prediction, of their activities. This study addresses this gap by analyzing data from Nigerian Security Tracker website (https://www.cfr.org/nigeria/nigeria-security-tracker/p29483) from year 2011 to 2023 and geolocated information on Boko Haram activity. Methods The research employs a mixed-methods approach. It uses descriptive statistics to understand attack trends and time series models (ARIMA/SARIMA) to forecast future attacks. Additionally, control charts identify periods of heightened insurgency. Results The findings confirm the Northeast region, as the epicenter of Boko Haram activities. The average monthly attack rate was 18 incidents, leading to 682 deaths over 12 years. 2014 and 2015 witnessed the peak of the insurgency. The forecasting models suggest a potential decrease in attack frequency in the coming years, with an average of nine attacks per month. This predicted decline might be linked to intervention efforts. Control charts reveal periods where attacks surpassed expected levels, highlighting critical moments for intensified counter-insurgency measures. These periods include July 2012-May 2014 and June 2014-August 2015, with a period of regained control. Conclusion This research provides valuable insights for stakeholders working to fight against Boko Haram’s insurgency. It offers forecasting capabilities and identifies critical periods, potentially informing targeted interventions and improving overall counter-insurgency strategies.

Intelligent hybrid deep learning models for enhanced shipboard solar irradiance prediction and charging station

Electrifying marine routes with solar energy significantly reduces ocean carbon emissions, playing a vital role in climate regulation. Solar irradiance forecasting ensures reliable power despite unpredictable sea weather, necessitating innovative model development. This research presents a forecasting model designed for the optimal placement of solar charging stations, providing hour-ahead solar irradiance predictions for onboard solar vessels. India (Jawaharlal Nehru Port Trust)– United States (Port of New York and New Jersey) sea route is selected to test our proposed forecasting model, as it aligns with busy shipping operations and long-distance trade requirements for electrification. Two distinct procedures by two case studies are used to select optimum data to predict the suitable locations for solar charging stations along the navigation route. In the first case study (Medium data analysis – MDA), the data are selected based on a statistical analysis of twenty years of hourly solar irradiance data from 201 locations. In the second case study (Large data analysis – LDA), two years of seasonal data from 201 locations are chosen and combined to form a large amount of hourly data. For both case studies, a hybrid solar irradiance forecasting model (ARIMA – Bi – LSTM) is developed by integrating the deep learning model with the time series model and fine-tuning them using optimization algorithms (PSO and GA). In MDA, the PSO optimization achieves mean absolute error (MAE) values of 0.32, 0.85, 1.83, 1.40, and 1.33 W/m2 for locations 1, 2, 3, 4, and 5, respectively, significantly outperforming the GA, which has MAE values of 1.36, 1.65, 3.25, 0.92, and 1.41 W/m2 for the same locations. In LDA, the PSO model achieves MAE values of 0.79, 0.91, 0.28, and 0.39 W/m2 for winter, rain, summer, and spring, respectively, also outperforming the GA, which has MAE values of 1.54, 2.17, 0.99, and 1.46 W/m2 for the same seasons. Comparatively, the LDA reinforces its best alignment to predict optimal charging station locations.

Temporal online self-learning stochastic configuration networks: A study on strip deviation prediction

Stochastic configuration networks (SCNs), with the rapid learning abilities and universal approximation properties, have garnered growing interest in the industrial domain in recent years. In the hot rolling process, strip deviation affects product quality and production safety. In this paper, a novel model named temporal online self-learning stochastic configuration networks (TOSL-SCNs) designed for early prediction of strip deviation in the hot rolling process is introduced. To address the challenges of time series prediction, conventional SCNs are extended to a two-dimensional format, enabling effective processing of time series data. Given the dynamic nature of industrial environments, the model incorporates an online self-learning capability to adapt to nonstationary data. Additionally, to ensure the stability of the model’s predictive performance, a data buffer strategy is introduced to dynamically refresh the training dataset. This paper demonstrates the capabilities of TOSL-SCNs in time series modeling, online self-learning, and data buffer strategy from multiple perspectives. The case study on strip deviation prediction highlights the necessity of these capabilities in industrial applications.

Latent level correlation modeling of multivariate discrete-valued financial time series

Latent level correlation modeling of multivariate discrete-valued financial time series

Optimize individualized energy delivery for septic patients using predictive deep learning models.

We aim to establish deep learning models to optimize the individualized energy delivery for septic patients. We conducted a study of adult septic patients in ICU, collecting 47 indicators for 14 days. We filtered out nutrition-related features and divided the data into datasets according to the three metabolic phases proposed by ESPEN: acute early, acute late, and rehabilitation. We then established optimal energy target models for each phase using deep learning and conducted external validation. A total of 179 patients in training dataset and 98 patients in external validation dataset were included in this study, and total data size was 3115 elements. The age, weight and BMI of the patients were 63.05 (95%CI 60.42-65.68), 61.31(95%CI 59.62-63.00) and 22.70 (95%CI 22.21-23.19), respectively. And 26.0% (72) of the patients were female. The models indicated that the optimal energy targets in the three phases were 900kcal/d, 2300kcal/d, and 2000kcal/d, respectively. Excessive energy intake increased mortality rapidly in the early period of the acute phase. Insufficient energy in the late period of the acute phase significantly raised the mortality as well. For the rehabilitation phase, too much or too little energy delivery were both associated with elevated death risk. Our study established time-series prediction models for septic patients to optimize energy delivery in the ICU. We recommended permissive underfeeding only in the early acute phase. Later, increased energy intake may improve survival and settle energy debts caused by underfeeding.

MPDDF: Multi perspective dual data flow model for time series forecasting

Abstract Multivariate time series forecasting tasks are widely used in social production and daily life, and making accurate predictions for the future can help enterprises take preventive measures for upcoming events. The prediction method based on deep neural networks has become the main method used in time series forecasting tasks. In order to improve the performance of the prediction model, this paper designs methods to improve the prediction performance of the model from the perspectives of multi-dimensional, time-domain, and data distribution in multivariate time series data. Using graph neural networks to establish dependencies between multi-dimensional data and perform feature extraction, using dual flow learning networks for time-domain feature extraction, and preventing overfitting of the model during training, and using learnable bias values to mitigate the impact of time series data distribution shift. By conducting experiments on the model in five different application scenarios and comparing it with advanced time series forecasting models, an average improvement of 7.50% was achieved in the MSE index and 6.53% in the MAE index, demonstrating the effectiveness of the model designed in this paper.

An IMFO-LSTM_BIGRU combined network for long-term multiple battery states prediction for electric vehicles

Predicting the state of power batteries is crucial for ensuring the safe and reliable operation of electric vehicles. However, accurately forecasting multiple battery parameters under full operating conditions remains challenging. Traditional prediction models struggle to accurately represent the actual battery behavior due to irregular vehicle driving patterns. Furthermore, long-term prediction of battery states, essential for potential fault diagnosis, poses significant difficulties for conventional neural networks. To address the issue of error accumulation in recurrent neural networks (RNNs) during multi-parameter battery prediction across various operating conditions, we propose a combined model featuring multi-level feature extraction. This model integrates Long Short-Term Memory (LSTM) and Bidirectional Gated Recurrent Unit (BiGRU) networks, where the LSTM layer employs forget gates to filter out unnecessary data while preserving valuable information. The bidirectional update gate in the BiGRU layer effectively combines historical and future data to enhance time series modeling. This comprehensive approach improves sequence modeling efficiency, thereby increasing reliability. However, the multi-level feature extraction structure introduces higher-dimensional hyperparameters, complicating the optimization process. To maintain prediction accuracy across multiple battery parameters, we propose an improved moth-flame optimization (IMFO) algorithm to optimize the complex hyperparameters of the LSTM_BiGRU model. Extensive experiments conducted on a real-world vehicle dataset demonstrate that the IMFO-LSTM_BiGRU combined network surpasses existing state-of-the-art methods in terms of accuracy and stability.

Relative position matrix and multi-scale feature fusion for writer-independent online signature verification

Online signature verification (OSV) is widely used in finance, law and other fields, and is one of the important research projects on biological characteristics. However, its data set has a small scale and has high requirements for generalization of certification models. Therefore, how to overcome these problems is of great value to improve the practicality and security of online handwriting signature technology. We propose a writer-independent online handwritten signature verification method, which adopts the relative position matrix method to convert the traditional temporal features into images for processing. This method enriched the features of the signatures, serving the purpose of data augmentation. Then two-dimensional multi-scale feature fusion based Siamese neural network (2D-MFFnet) is built for representing and learning the importance of each channel adaptively combined with the attention mechanism. Finally, a temporal convolutional network is designed to construct the classifier. The results illustrate that compared with traditional time series models, the algorithm has reduced the equal error rate by at least 2.52 % on the open datasets MCYT-100 and SVC2004 task2.

Research on Sales of New Energy Vehicles Based on ARIMA Model

Firstly, this article identifies seven indicators based on electric vehicles: battery endurance, number of public charging facilities, government subsidies, gasoline prices, electricity prices, carbon emissions, and number of power battery companies. Spearman correlation analysis is conducted between indicator data and sales of new energy electric vehicles. The correlation coefficients of the first five indicators are all greater than 0.6, which is the main influencing factor. Then, the sales of new energy vehicles are predicted using a yearly forecasting method. Due to the close correlation between the sales of new energy vehicles and the five main indicators, the ARIMA time series model is first used to predict the annual quantity of the five indicators. Then, using these five indicators as inputs and sales as outputs, a BP neural network model is established to predict the annual sales of new energy vehicles in China to be 46051748 units in ten years. Finally, quantifying the score of ecological environment assessment using four evaluation indicators: sulfur dioxide, smoke emissions, ammonia nitrogen, and carbon dioxide. The entropy weight method was used to determine the weights of each indicator, and the results were 37.25%, 30.96%, 21.24%, and 10.55%, respectively. Then apply it to the TOPSIS model to obtain ecological environment scores for different years, and analyze the correlation between the degree of electrification and ecological environment scores. Finally, it can be concluded that the larger the market share of new energy vehicles, the better the ecological environment.

Effect Analysis of Covid-19 on China’s Capital Market for Property-Rights Exchange Based on Functional Clustering

In August 2015, the State Council of the People’s Republic of China listed China’s capital market for property-rights exchange as the capital market along with the Chinese stock market. Consequently, China’s capital market for property-rights exchange is an integral part of the country’s multilayered capital market. The COVID-19 pandemic has had significant effects on the capital markets. Since China’s capital market for property-rights exchange mainly serves nonlisted companies, it is complicated to obtain its exchange data to perform COVID-19’s effect analysis. This study investigates COVID-19’s effect on China’s capital market for property-rights exchange. In this paper, we collected the online property rights-exchange data from 2017 to 2020 of an electric trading platform. The functional principal component analysis method is innovatively introduced to explore the online property rights-exchange fluctuation to get an insight into COVID-19’s effect on China’s capital market for property-rights exchange. Based on the principal component scores, COVID-19’s effect on different property rights-exchange institutions are divided into two categories using principal component clustering. A time-series model is used to quantify the effects in the two categories. The research results finally show that online property rights-exchange can reduce the negative effect of COVID-19 on China’s capital market for property-rights exchange, which encourages property rights-exchange institutions to accelerate the use of the online platform to overcome the challenges of the pandemic.