Time Series Analysis Methods Research Articles

Multilayer LSTM Model for Wind Power Estimation in the Scada System

Wind energy is clean energy that does not pollute the environment. However, the complex and variable operating environment of a wind turbine often makes it difficult to predict the instantaneous active power generated. In this study, a wind turbine active power estimation system based on a short-term memory network (LSTM) using time series analysis is proposed. The data obtained from the wind turbine SCADA system is used as input variables. In the proposed method, a multilayer LSTM architecture is designed to train the model. The first LSTM network consists of 64 units, and the second one consists of 32 units. This is followed by a dense layer consisting of 16 neurons. In the last layer, the architecture is finalized by using a linear activation function for the prediction process. The proposed deep learning (DL)-based LSTM prediction model takes into account environmental factors such as wind speed and wind direction for active power forecasting. The results show that the LSTM-based time series analysis method is capable of effectively capturing time series features among the data. Thus, the proposed architecture can realize high-accuracy active power forecasting.

Time Series and Cyclical Analysis on Chinese GDP Growth in 2003-2022

This paper analyzes the trend of China's GDP from 2003 to 2022 using time series analysis, detrending and business cycle method. It highlights the importance of government investment in infrastructure and human capital to enhance efficiency in the Chinese economy. The paper finds that consumption, investment, and government consumption are highly positively correlated with GDP, suggesting that the government can stimulate the economy by increasing mass consumption, business investment, and government purchases. However, the value of net exports, which depends on the simultaneous changes of export value and import value, has great uncertainty and makes it difficult for the government to stimulate the economy through changes in net exports. Furthermore, the paper suggests that the priority in stimulating the economy should be given to increasing government consumption, followed by consumption and investment. The paper proposes several stimulus measures, including increasing infrastructure construction, issuing consumer vouchers, and reducing interest rates for business loans. These measures aim to boost government consumption, stimulate consumption, and encourage business investment, ultimately contributing to economic growth and development.

Statistical Analysis of the Socio-Economic Situation in St. Petersburg in the Context of Crises

Purpose of the study. Statistical research of the socio-economic development issues of the region acquires special significance in crisis situations. Statistical methods make it possible to quantify the current situation, identify the impact of crisis phenomena and forecast the development of the region. In the analysis of socio-economic development, both economic characteristics (GDP, production and distribution) and direct indicators of the standard of living of the population (unemployment level, personal income and wages) are the key ones. The purpose of this paper is to statistically assess the impact of crisis situations on the socio-economic situation in St. Petersburg.Materials and methods. The information base for the study was official data from the Federal State Statistics Service. The paper uses methods of time series analysis, cluster analysis and analysis of structural changes.Results. Based on the results of the study, the main periods in the dynamics of the main socio-economic indexes were identified. Special attention is paid to studying the structure of GRP of St. Petersburg. Emphasis is placed on comparing the dynamics of regional indexes with data for the federal district and the Russian Federation as a whole. The features of the city’s development are considered using cluster analysis according to the current range of factors for different time periods. An assessment of the achievement of key planned indexes is given and a comprehensive index is proposed for assessing the development of the dynamics of the socio-economic development of the region. Particular attention is paid to analyzing the impact of crisis situations on the socio-economic status of the region and assessing the degree of their impact.Conclusion. Based on the results of the analysis, recommendations were given for the development of the region, which can be used in the implementation of socio-economic policy. Particular attention is paid to indexes of living standards.

Study on the Variation in Coastal Groundwater Levels under High-Intensity Brine Extraction Conditions

The excessive exploitation of groundwater is becoming a serious global issue. Different from other regions, groundwater extraction in coastal areas usually stops and moves inland after causing seawater intrusion. The abundant salt fields in the Laizhou Bay area of China provide a unique case of maintaining high-intensity underground brine mining even after seawater intrusion. The intensive exploitation of underground brine has led to significant changes in the groundwater flow field. However, there is still a lack of research on how different factors affect the groundwater level in this mining situation. In this paper, time series analysis methods were used to investigate the impact of brine water extraction, tidal fluctuations, and precipitation on the groundwater level in the Laizhou Bay area. The results indicate that brine extraction is the main factor controlling the changes in groundwater level, with the cessation and resumption of extraction resulting in a 93.4 cm increase and a 122.5 cm decrease, respectively. Different rainfall patterns can also lead to an increase in groundwater levels, especially when a heavy rainfall event can cause a 61.2 cm increase. Tidal fluctuations can cause periodic fluctuations in the groundwater level, with a variation amplitude of approximately 11% of the tide itself.

Impact of the COVID-19 pandemic on antibiotic prescribing in high-prescribing primary care physicians in Switzerland

ObjectivesThe aim of this study was to evaluate antibiotic prescribing of medium-to-high prescribing primary care physicians being followed up after the completion of a Swiss national intervention trial of antibiotic prescription audit and feedback in the first SARS-CoV-2 pandemic year. MethodsWe used health insurer based claims data to calculate monthly antibiotic prescription rates per 100 consultations (primary endpoint) and applying interrupted time series (ITS) analysis methods, we estimated the immediate (step change) and sustained effects (slope) of the SARS-CoV-2 epidemic in 2020 on antibiotic prescribing compared to the pre-pandemic trial period from 2017–2019. ResultsWe analysed data of 2945 of 3426 physicians (86.0%) from the trial with over 4 million consultations annually, who were in 2020 still in practice. Consultations dropped by 43% during the first pandemic year compared with 2017. Median monthly antibiotic prescription rates per 100 consultations in 2017 were 8.44 (Interquartile range [IQ] 6.32–11.50) and 8.35 (6.34–11.74) in the intervention and control groups, respectively, and increased to 15.63 (10.69–23.81) and 16.31 (10.65–24.72) per 100 consultations in 2020. ITS-derived incidence rate ratios for overall antibiotic prescriptions were 2.32 (95% CI 2.07–2.59) for the immediate pandemic effect, and 0.96 (0.95–0.98) for the sustained effect (change in slope in 2020 compared with 2017–2019). DiscussionThe SARS-CoV-2 pandemic had a major impact on antibiotic prescription patterns in primary care in Switzerland. For future viral pandemics, intervention plans with timely activation steps to minimize unjustified antibiotic consumption in primary care should be prepared.

Strategic Performance Management Based on Cigarette Production and Operation Big Data

Abstract The traditional strategic performance management of cigarette enterprises is not sufficiently comprehensive and objective. This paper constructs a big data model of cigarette production and sales by collecting production and operation data, using the primary sliding average method of time series analysis to forecast the short-term sales of cigarettes, and combining it with the coefficient of the dynamic sales ratio. The operational big data is applied to strategic performance management under the balanced scorecard, and the performance management index system is constructed with strategic orientation in mind. Based on the production and sales data of Cigarette Factory S, the weights of the indicators are determined and applied to the empirical study. The scores of the indicators of finance, customer, internal operation learning and growth of Cigarette Factory S are 78.14, 82.29, 87.07 and 81.68, respectively. The performance management system of the Cigarette Factory under the strategic orientation is feasible.

Abstract 16676: Risk Assesment of Sudden Cardiac Death With RR Interval Fluctuations

Introduction: Changes in conventional measures for short-term heart rate variability (HRV) have been associated with heart failure, coronary artery disease and even sudden cardiac death (SCD). However, the predictive value of HRV in clinical settings has been limited. Secondly, the current HRV methods have not discriminated SCDs from other causes of death, which would be relevant for proper prevention and treatment. In this study we utilize novel time series analysis methods for RR intervals in ultra-short-term measurements to distinguish SCD from other causes of death. Methods: We apply second-order detrended fluctuation analysis (DFA) with maximally overlapping windows to assess the correlations in RR intervals as a function of the scale, i.e., the range of consecutive RR intervals in one-minute resting ECG. The analysis is based on the baseline ECG recordings from the prospective FINCAVAS study of cardiac patients with a median follow-up time of 8.3 years, during which 378 deaths were recorded including 76 SCDs in the total population of 2794 subjects. Results: Our DFA method can discriminate between incident SCD cases and patients dying to other causes. The discrimination is maximized on scale 9, with an univariate hazard ratio of 0.71 and 95% confidence intervals (CI) (0.55-0.91). Excluding other cardiac deaths from the control group we are left with 279 deaths resulting to univariate hazard ratio of 0.62 (0.47-0.82 CI) on scale 8. Conclusions: The ability to distinguish SCD from other causes of death shows that the scale-dependent DFA method is a superior alternative to the conventional HRV methods. The utilization of one-minute RR interval measurements for risk assessment provides new non-invasive tools for clinical inspections, as well as promising prospects to health monitoring with wearable devices.

Casino-based cash transfers and fertility among the Eastern Band of Cherokee Indians in North Carolina: A time-series analysis

Fertility decline remains a key concern among high-income countries. Prior research indicates that income supplementation through unconditional cash transfers (UCT) may correspond with increased fertility. We examine whether a casino-based UCT, in the form of per capita (percap) payments to members of the Eastern Band of Cherokee Indians (EBCI) corresponds with an acute increase in fertility. We use North Carolina vital statistics datasets from 1990 to 2006 and apply time-series analysis methods to examine the relation between specific months of percap payments (exposure) and monthly number of conceptions that result in live births (outcome) among the EBCI. We control for autocorrelation and monthly counts of births (arrayed by conception cohorts) among white women (ineligible for UCT receipt) in the study region. Results indicate an increase in conceptions that result in live births at 1 and 3 months after percap receipt among EBCI women aged ≥20 years (exposure month lag 1 coefficient = 1.74, p = 0.03; exposure month lag 3 coefficient = 1.60, p = 0.04). Exploratory analyses indicate that the observed fertility increase concentrates among primiparae EBCI women. We do not find any association between percap payment timing and births to EBCI women aged <20 years.

A new network representation for time series analysis from the perspective of combinatorial property of ordinal patterns

Revealing system behavior from observed time series is a fundamental problem worthy of in-depth study and exploration, and has attracted extensive attention in a wide range of fields due to its wide application values. In this paper, we propose a novel network construction method for time series analysis, which is different from the existing ordinal network method concerning the transition probability of ordinal patterns in transition networks. The proposed network representation is based on the combinatorial property concerning the inversion number of ordinal patterns from the ordinal partitions of time series. For the proposed network construction method, the network nodes are represented by each ordinal partition of time series and the edge weight between network nodes is determined by a novel proximity relationship of ordinal patterns which is a newly defined metric based on the inversion number of ordinal patterns. Using random signals and chaotic signals as examples, we demonstrate the potential of the proposed network construction method for the network representation of time series. We also employ the proposed network construction method in quantitative EEG for the identification of three different physiological and pathological brain states. According to the results of AUC values, one can observe that the discriminating power of the AND of the proposed network construction method is slightly stronger than that of the available ordinal network. The experimental results illustrate that our proposed network construction method opens up a new pathway for network representation of time series, which is capable of quantifying time series for feature extraction and pattern learning for time series analysis.

Comparison of prediction methods for added power and propeller revolution for a 1,800 TEU container ship using resistance tests in regular and irregular waves

This paper predicts the added power and propeller revolution due to irregular head waves for a 1,800 TEU container ship. Resistance tests were conducted with 11 regular waves and irregular waves at SS = 5. Time series (TS) and autocovariance analysis methods were applied to estimate statistical variables and uncertainties. The paper investigated the statistical convergence times of measured data. The wave elevation, heave and pitch motions obtained from the TS agree well with those from spectral analysis in regular and irregular waves. The added resistance from the TS is larger than that from spectral analysis using a regular wave test by 6.1%. The characteristics of response operators and spectra of the added (or reduced) propulsion performances and overload factors are discussed. The power and propeller revolution were evaluated with a variety of methods: SA(LVM-W), LVM-IR, DPM, and RTIM.

LPI Radar Detection Based on Deep Learning Approach with Periodic Autocorrelation Function.

In electronic warfare systems, detecting low-probability-of-intercept (LPI) radar signals poses a significant challenge due to the signal power being lower than the noise power. Techniques using statistical or deep learning models have been proposed for detecting low-power signals. However, as these methods overlook the inherent characteristics of radar signals, they possess limitations in radar signal detection performance. We introduce a deep learning-based detection model that capitalizes on the periodicity characteristic of radar signals. The periodic autocorrelation function (PACF) is an effective time-series data analysis method to capture the pulse repetition characteristic in the intercepted signal. Our detection model extracts radar signal features from PACF and then detects the signal using a neural network employing long short-term memory to effectively process time-series features. The simulation results show that our detection model outperforms existing deep learning-based models that use conventional autocorrelation function or spectrogram as an input. Furthermore, the robust feature extraction technique allows our proposed model to achieve high performance even with a shallow neural network architecture and provides a lighter model than existing models.

The popularity function: a spurious regression? The case of Austria

AbstractIn this paper we apply the unit root and cointegration methodology as well as other methods of modern econometric time series analysis to estimate popularity functions for the Austrian parties in power since the mid-1970s. We find only very rare evidence for economic variables influencing the popularity of the main political parties in the federal government, thereby challenging previous studies that claimed to have established such influences. When considering the determinants of the popularity of the incumbent party (the sum of the popularities of the parties in government), the unemployment rate turns out to have a negative short-run influence. The robustness of this result is checked by various methods, demonstrating the usefulness of the applied methodology to clarify such issues.

Characterization of homogeneous nucleation of superheated liquid droplets by nonlinear analysis

The homogeneous bubble nucleation in superheated liquid droplets can be initiated by energetic particles or radiation. The nonlinearity in the heat transfer introduces chaos in the temperature time series of a boiling system. In the present study, the acoustical pulses from the homogeneous nucleation of superheated liquid droplets by neutrons and gamma-rays have been collected. Nonlinear time series analysis methods like 3D attractor reconstruction and Fast Fourier transformation (FFT) have been performed on the collected pulses. The analysis predicts the possibility of the existence of chaos, and the positive value of the largest Lyapunov exponent confirms it. The nonlinearity in the energy deposition by both the recoil nuclei and electrons from the neutrons and gamma-rays respectively, introduce chaos. A qualitative difference has been observed in the 3D attractor and Rescaled Range (R/S) plots between the neutron and gamma-ray-induced pulses, and the present differences are useful for the identification of these pulses.

Weighted Dynamic Time Warping for Time Series

Recurrence network is a typical time series analysis method. However, irregular sampling may overshadow the dynamic features characterized by traditional recurrence network method, which makes the method ineffective. This paper introduces dynamic time warping method to determine the distance between time series segments. The method can match the features of time series segments and reduce the distortion of distances caused by irregular sampling. For points in sparse sampling area, many characteristics may not be collected and thus the matching results are unreliable. To overcome this problem, we further propose weighted dynamic time warping method which assigns small weights to matching points with sparse local sampling so as to reduce their influence on distance. A recurrence network constructed by weighted dynamic time warping method can effectively capture underlying nonlinear and nonstationary system dynamics from irregular sampling time series. The effectiveness of the proposed method is illustrated by two case studies: the discrimination of different dynamic behaviors and detection of system parameter perturbation from time series of Rössler system.

Leveraging PyCaret for Time Series Analysis-A Low Code Approach

Time series analysis is a fundamental task in various domains, ranging from finance to healthcare and beyond. Traditional methods for time series analysis often require significant manual effort and expertise. PyCaret, a low-code machine learning library, offers a simplified approach to time series analysis, enabling practitioners to build robust models with minimal code. In this paper, we delve into PyCaret’s capabilities for time series analysis, exploring its methods and comparing them with traditional Python packages. Through examples and case studies, we demonstrate how PyCaret streamlines the time series analysis workflow, making it accessible to a broader audience.

COMPARISON OF THE EFFECTIVENESS OF TIME SERIES ANALYSIS METHODS: SMA, WMA, EMA, EWMA, AND KALMAN FILTER FOR DATA ANALYSIS

In time series analysis, signal processing, and financial analysis, simple moving average (SMA), weighted moving average (WMA), exponential moving average (EMA), exponential weighted moving average (EWMA), and Kalman filter are widely used methods. Each method has its own strengths and weaknesses, and the choice of method depends on the specific application and data characteristics. It is important for researchers and practitioners to understand the properties and limitations of these methods in order to make informed decisions when analyzing time series data. This study investigates the effectiveness of time series analysis methods using data modeled with a known exponential function with overlaid random noise. This approach allows for control of the underlying trend in the data while introducing the variability characteristic of real-world data. The relationships were written using scripts for the construction of dependencies, and graphical interpretation of the results is provided.

Dynamic mode decomposition with memory.

This study proposed a numerical method of dynamic mode decomposition with memory (DMDm) to analyze multidimensional time-series data with memory effects. The memory effect is a widely observed phenomenon in physics and engineering and is considered to be the result of interactions between the system and environment. Dynamic mode decomposition (DMD) is a linear operation-based, data-driven method for multidimensional time-series data proposed in 2008. Although DMD is a successful method for time-series data analysis, it is based on ordinary differential equationsand thus cannot incorporate memory effects. In this study, we formulated the abstract algorithmic structure of DMDm and demonstrate its utility in overcoming the memoryless restriction imposed by existing DMD methods on the time-evolution model. In the numerical demonstration, we utilized the Caputo fractional differential to implement an example of DMDm such that the time-series data could be analyzed with power-law memory effects. Thus, we developed a fractional DMD, which is a DMD-based method with arbitrary (real value) order differential operations. The proposed method was applied to synthetic data from a set of fractional oscillators and model parameters were estimated successfully. The proposed method is expected to be useful for scientific applications and aid in model estimation, control, and failure detection of mechanical, thermal, and fluid systems in factory machines, such as modern semiconductor manufacturing equipment.

Analyzing and Forecasting Pediatric Fever Clinic Visits in High Frequency Using Ensemble Time-Series Methods After the COVID-19 Pandemic in Hangzhou, China: Retrospective Study.

The COVID-19 pandemic has significantly altered the global health and medical landscape. In response to the outbreak, Chinese hospitals have established 24-hour fever clinics to serve patients with COVID-19. The emergence of these clinics and the impact of successive epidemics have led to a surge in visits, placing pressure on hospital resource allocation and scheduling. Therefore, accurate prediction of outpatient visits is essential for informed decision-making in hospital management. Hourly visits to fever clinics can be characterized as a long-sequence time series in high frequency, which also exhibits distinct patterns due to the particularity of pediatric treatment behavior in an epidemic context. This study aimed to build models to forecast fever clinic visit with outstanding prediction accuracy and robust generalization in forecast horizons. In addition, this study hopes to provide a research paradigm for time-series forecasting problems, which involves an exploratory analysis revealing data patterns before model development. An exploratory analysis, including graphical analysis, autocorrelation analysis, and seasonal-trend decomposition, was conducted to reveal the seasonality and structural patterns of the retrospective fever clinic visit data. The data were found to exhibit multiseasonality and nonlinearity. On the basis of these results, an ensemble of time-series analysis methods, including individual models and their combinations, was validated on the data set. Root mean square error and mean absolute error were used as accuracy metrics, with the cross-validation of rolling forecasting origin conducted across different forecast horizons. Hybrid models generally outperformed individual models across most forecast horizons. A novel model combination, the hybrid neural network autoregressive (NNAR)-seasonal and trend decomposition using Loess forecasting (STLF), was identified as the optimal model for our forecasting task, with the best performance in all accuracy metrics (root mean square error=20.1, mean absolute error=14.3) for the 15-days-ahead forecasts and an overall advantage for forecast horizons that were 1 to 30 days ahead. Although forecast accuracy tends to decline with an increasing forecast horizon, the hybrid NNAR-STLF model is applicable for short-, medium-, and long-term forecasts owing to its ability to fit multiseasonality (captured by the STLF component) and nonlinearity (captured by the NNAR component). The model identified in this study is also applicable to hospitals in other regions with similar epidemic outpatient configurations or forecasting tasks whose data conform to long-sequence time series in high frequency exhibiting multiseasonal and nonlinear patterns. However, as external variables and disruptive events were not accounted for, the model performance declined slightly following changes in the COVID-19 containment policy in China. Future work may seek to improve accuracy by incorporating external variables that characterize moving events or other factors as well as by adding data from different organizations to enhance algorithm generalization.

Analysis of the development of energy consumption of electricity and fuels in Slovakia

As the standard of living rises, the energy consumption of electricity and fuels increases not only in Slovakia but across the European Union. Future energy demand will be based on population growth, and it is the key factor of economic and social development of the society, with an emphasis being put on improving the quality of life. This study focused on the development of the final energy consumption (FEC) of individual fuels, electricity and heat, including their shares in the final energy consumption in selected economy sectors in Slovakia. Time-series analysis methods were applied to predict energy consumption for the next three years. Although we have witnessed a decrease in the total energy consumption in individual sectors in Slovakia in the years 2005–2015, energy consumption generally exhibits an increasing trend. In all of the analysed sectors, the development of final consumption of energy from renewable energy sources (RESs) and wastes exhibited a rising trend; nevertheless, the results of our evaluation indicated that Slovakia still has some major shortcomings as to the utilisation of RESs

Discussion on the Advantages and Disadvantages of the DCF Model in the Current Era and the Improvement of the Model

The discounted cash flow model is a company value evaluation method widely used by many domestic investors when investing company. However, under the background of the epidemic, due to factors like low benchmark interest rates that are not conducive to the valuation of the DCF model, the accuracy of the discounted cash flow model valuation has been questioned. In addition, the DCF model is not entirely applicable to the domestic market because of the multiple systems and the local capital market's imperfections, so the model needs to be improved. In this paper, the existing literature is summarized, and it was found that the time series analysis method can be used, which can predict the company's future cash flow more accurately, instead of the sales ratio analysis method, while the discount rate can be calculated more accurately by substituting each company's beta into the calculation instead of using the company's industry beta.