Basis Of Time Series Research Articles

Electric vehicle load forecasting based on convolutional networks with attention mechanism and federated learning method

AbstractAccurate forecasting of electric vehicle (EV) load is essential for grid stability and energy management. EV load forecasting is influenced by multiple factors. At present, the load forecasting model for EVs mainly uses collected sample data to build a data‐driven model. But these algorithms need to collect all the data together to train the model, ignoring the privacy of each data collection source. In a competitive market environment, each device service provider is not willing to share the sample data they store. Aiming at this problem, this paper proposes an EV load diagnosis algorithm considering data privacy. Firstly, a convolutional neural network with dual attention mechanism is constructed as the basic time series forecasting model. The association rule algorithm is used to select weather data with strong associations as the inputs of the model. Each service provider uses local data to perform deep learning network. All models are then trained using a federated learning framework. During the entire training process, historical data is stored locally, and only model parameter information is shared and interacted; thus data privacy is protected. Finally, the validity of the algorithm in this paper is verified by using real collected EV load data.

The relationship between the Debt-to-GDP ratio and the GDP in developed (US, Japan), developing countries (of Asia and Europe) and African sub-saharan countries with emphasis on Indian scenario: A comparative study

There is no simple relationship between debt and growth […]. There are many factors that matter for a country’s growth and debt performance. Moreover, there is no single threshold for debt ratios that can delineate the “bad” from the “good”. (International Monetary Fund, 2012, p. 9)Gross Domestic Product (GDP) is often used as an indicator of the size of the economy and the debt-to-GDP ratio works as an indicator of the financial leverage for an economy. A low ratio points that an economy’s goods and services production is adequate to pay off its debts without letting further debts be incurred. The borrowing pattern of a nation and the election to opt to incur further debt depends on economic and geopolitical considerations which include recession, war, interest rates etc. On the other hand, a high ratio would imply that the economy is not producing sufficiently to pay off its debts. Just like any bank would be interested in providing a bigger amount of loan only when an individual makes more money; likewise, in an economy’s scenario, investors would be more interested in taking on a country’s debt if it could produce more. And if at any time investors happen to worry about the repayment, and then they start to ask for higher interest rate returns to secure themselves against the risk of default. This way, it increases the cost of debt and the economy might fall into the trap of debt crisis.This paper investigates the impact of India's public debt on its economic growth through an econometric analysis using data from the Reserve Bank of India, the International Monetary Funds, and the World Development Indicators for the period 1989-2014. The data is regressed in basic time series analysis taking into account the different variables that influence economic growth. The regression results show an inverted U-shape relationship between the public debt to GDP and its square. The results illustrate the theoretical findings of Reinhart and Rogoff's (2010) changing relationship between real GDP growth and government debt based on a debt threshold.

Financial assets prediction based on ARIMA, Random Forest and GRU

As a matter of fact, financial asset prediction is a domain of great interest because of its potential to generate revenue. In reality, financial asset prediction models have evolved from basic time series analysis models to contemporarily hybrid models with the help of machine learning algorithms. To be specific, this study will introduce and analyse three popular financial asset forecasting models and their hybrids in terms of the properties they have demonstrated in the completed studies. In reality, good results have been achieved using a time series analysis method named autoregressive integrated moving average (ARIMA) for capturing linear elements. According to the analysis, in dealing with data noise and interpretability, Random Forest (RF) algorithm, a machine learning technique, produced positive outcomes. Deep learning technique gate recurrent unit (GRU) produced positive outcomes in terms of prediction accuracy. Based on the evaluations, this study indicates future research directions in the field of financial asset forecasting by analysing and organizing the characteristics of three different mainstream models.

Using Time Series Analysis to Study the Changing Trend of Business English Teaching Effectiveness in Colleges and Universities

Abstract This paper analyzes the characteristics and changes of influence on college business English teaching in the context of the new era, constructs a multimodal college business English teaching model based on vision and hearing, and carries out a multimodal fusion of graphics, text and audio through deep learning. To solve this problem, the time series analysis problem is easily affected by noise. On the basis of the basic time series algorithm, the GBDT-KF algorithm based on the Kalman filter is designed. It also predicts the effectiveness of college business English teaching. The results show that listening ability is improved by 0.3, speaking ability by 0.25, writing ability by 0.21, and reading ability by 0.18; the effect of college business English teaching has also been improved in different degrees.

The soft computing FFNN method for adjusting heteroscedasticity on the time series model of currency exchange rate

IntroductionTime series models on financial data often have problems with the stationary assumption of variance on the residuals. It is well known as the heteroscedasticity effect. The heteroscedasticity is represented by a nonconstant value that varies over time.MethodsThe heteroscedasticity effect contained in the basic classical time series model of Autoregressive Integrated Moving Average (ARIMA) can adjust its residuals as the variance model by using Generalized Autoregressive Conditional Heteroscedasticity (GARCH). In improving the model accuracy and overcoming the heteroscedasticity problems, it is proposed a combination model of ARIMA and Feed-Forward Neural Network (FFNN), namely ARIMA-FFNN. The model is built by applying the soft computing method of FFNN to replace the variance model. This soft computing approach is one of the numerical methods that can not be only applied in the theoretical subject but also in the data processing.ResultsIn this research, the accuracy of the time series model using the case study of the exchange rate United States dollar-Indonesia rupiah with a monthly period from January 2001 to May 2021 shows that the best accuracy of the possible models is the model of ARIMA-FFNN, which applies soft computing to obtain the optimal fitted parameters precisely.DiscussionThis result indicates that the ARIMA-FFNN model is better used to approach this exchange rate than the rest model of ARIMA-GARCH and ARIMA-GARCH-FFNN.

A Basic Time Series Forecasting Course with Python

The aim of this paper is to present a set of Python-based tools to develop forecasts using time series data sets. The material is based on a 4-week course that the author has taught for 7 years to students on operations research, management science, analytics, and statistics 1-year MSc programmes. However, it can easily be adapted to various other audiences, including executive management or some undergraduate programmes. No particular knowledge of Python is required to use this material. Nevertheless, we assume a good level of familiarity with standard statistical forecasting methods such as exponential smoothing, autoregressive integrated moving average (ARIMA), and regression-based techniques, which is required to deliver such a course. Access to relevant data, codes, and lecture notes, which serve as based for this material, is made available (see https://github.com/abzemkoho/forecasting) for anyone interested in teaching such a course or developing some familiarity with the mathematical background of relevant methods and tools.

Maximum likelihood estimation for uncertain autoregressive moving average model with application in financial market

Uncertain time series analysis is an influential component of statistics that employs chronological data for further application in forecasting and control. As basic time series models, the uncertain autoregressive model and uncertain moving average model can not deal with the situation where the current observation is impacted by both the past observations and the past disturbance terms. This motivates us to initiate an uncertain autoregressive moving average model for obtaining better flexibility and general ability in actual problems. First, this paper presents a maximum likelihood estimation for calculating the parameters of the uncertain autoregressive moving average model and defines a mean absolute deviation criterion to identify it. This procedure is applied to the range return of the financial market, which is obtained from low and high prices of the corresponding trading day. Then, two examples of gold futures price and Microsoft stock price are applied to illustrate the accuracy of this method. The uncertain hypothesis test is used to analyze the properties of the residuals and correct for outliers. Finally, two comparative analyses are given to show the robustness of maximum likelihood estimation in the presence of outliers and the need to introduce the uncertain autoregressive moving average model.

Prediction on the Value Trends of Bitcoin and Gold-on Account of ARMA Time Series Forecasting Model

In this paper, we aimed to build a quantitative investment trading model based on a combination of a multivariate cycle ARMA model and Apriori. We first note that in order to have a sound investment strategy, a forecast for the next trading day needs to be made. To do this, a basic time series forecasting model was first built to predict the value of gold and bitcoin for the next day based on the market volatility of the previous 40 days. The next step is developing a trading strategy model with a stable rate of return and some risk tolerance. At the same time, we developed a fixed stop-loss strategy to protect the strategy's stability and improve the risk resistance performance. Ultimately, using this model, we calculated that on 10 September 2021, we will have a return of $4816941 in Bitcoin and $1129.0503 in gold.

Is the Spatial-Temporal Dependence Model Reliable for the Short-Term Freight Volume Forecast of Inland Ports? A Case Study of the Yangtze River, China

The purpose of this study is to investigate whether spatial-temporal dependence models can improve the prediction performance of short-term freight volume forecasts in inland ports. To evaluate the effectiveness of spatial-temporal dependence forecasting, the basic time series forecasting models for use in our comparison were first built based on an autoregression integrated moving average model (ARIMA), a back-propagation neural network (BPNN), and support vector regression (SVR). Subsequently, combining a gradient boosting decision tree (GBDT) with SVR, an SVR-GBDT model for spatial-temporal dependence forecast was constructed. The SVR model was only used to build a spatial-temporal dependence forecasting model, which does not distinguish spatial and temporal information but instead takes them as data features. Taking inland ports in the Yangtze River as an example, the results indicated that the ports’ weekly freight volumes had a higher autocorrelation with the previous 1–3 weeks, and the Pearson correlation values of the ports’ weekly cargo volume were mainly located in the interval (0.2–0.5). In addition, the weekly freight volumes of the inland ports were higher depending on their past data, and the spatial-temporal dependence model improved the performance of the weekly freight volume forecasts for the inland river. This study may help to (1) reveal the significance of spatial correlation factors in ports’ short-term freight volume predictions, (2) develop prediction models for inland ports, and (3) improve the planning and operation of port entities.

Baseline accuracy of forecasting COVID-19 cases in Moscow region on a year in retrospect using basic statistical and machine learning methods

The large amount of data that has accumulated so far on the dynamics of the COVID-19 outbreak has allowed to assess the accuracy of forecasting methods in retrospect. This work is devoted to comparing a set of basic time series analysis methods for forecasting the number of confirmed cases for 14 days ahead: machine learning methods, exponential smoothing, autoregressive methods, along with variants of SIR and SEIR. On the year-long data for Moscow, the best basic model is showed to be SEIR within which the basic reproduction number R0 is predicted using a regression model, achieving the mean error of 16% by the MAPE metric. The resulting accuracy can be considered a baseline for a more complex prospective model that would be based on the presented approach.

Prospects for innovation performance on European level

Abstract In 2004, the European Commission implemented the Decision No 1608/2003/EC of the European Parliament and of the Council concerning the production and development of Community statistics on innovation. This triggered the awareness of the role of innovation and R&D on national and European level and thus the opportunity to step towards in-depth monitoring innovation performance through various indicators. The paper aims to investigate the trends in the selected innovation indicators (i.e., public funding, expenditures and innovation activities, types of innovation and products introduced, hampered innovation activities) to outline the development direction on the enterprise level using the Community innovation survey data for the 2002–2016 period. Using the basic time series analysis, the paper evaluates the progress according to the European Strategy on research and innovation. Furthermore, using the autocorrelation and autoregression methods, the paper also outlines the future direction in innovation performance on European level.

Occupational Disease as the Bane of Workers’ Lives: A Chronological Review of the Literature and Study of Its Development in Slovakia. Part 1

This article not only offers a chronological overview of the development of occupational medicine, but also offers a summary of occupational diseases recommended by the ILO and legislative decisions that have influenced how we approach assessment today. We consider that these areas form a whole in which they cannot function without each other and they would lose their relevance if the system was collapsed. By excluding even one part of it, we would find ourselves at the beginning of the era of occupational medicine, and a large number of employees would once again be exposed to conditions that previously led to considerable illness and mortality of employees. The article also examines legislation and the development of occupational diseases in Slovakia in the period 1997–2019. Using basic statistical methods and time series, a trend model for the time series of the development of the number of occupational diseases over the last 20 years is created. The modeling also includes a forecast for the development of the number of occupational diseases for the next 5 years. The model created shows a favorable, decreasing trend in the number of occupational diseases in Slovakia.

Suppressing Black Votes: A Historical Case Study of Voting Restrictions in Louisiana

Southern states have used a variety of methods to disenfranchise African American voters. Empirical data on the effectiveness of these measures is rare. We present a unique data source from Louisiana that allows us to empirically document voter registration rates from the end of Reconstruction to the present. Using basic time series data, we document how voter registration rates changed over time in response to state restrictions. We then conduct a second analysis, which focuses on Louisiana’s use of the Understanding Clause to reduce voter registration among Blacks. We show that in parishes that used the Understanding Clause, Black registration rates dropped by nearly 30 percentage points, with little effect on white registration. The findings of this paper have important implications for understanding the potential for discrimination in the enforcement of modern, ostensibly nonracial, voter eligibility requirements, such as voter ID laws, which grant substantial discretion to local officials in determining voter eligibility.

Basic Time Series Models in Financial Forecasting

Forecasting is inevitable process of modern day life. It is about predictions of the future based on a set of historical or present information and most commonly accompanied by analysis of trends. Forecasting employs different forecasting techniques depending on the source of information and the objective of forecast. The main interest of this article are the basic time series methods used in financial forecasting, in particular, simple moving average method, weighted moving average method and the exponential moving average method. A version of the simple moving average is the cumulative moving average, which is also presented in a short term. These are generally used to forecast time series without trend manifestation or seasonal component. The analysis of the forecasting methods include reviews of the basic forecasting concepts, suggestions for possible application and comparison of the featuring limitations.

Seasonality in Covid-19 Times

COVID-19 has hit the economy in an unprecedented way, abruptly changing the data generating process of many economic time series. This has triggered one of the highest policy interventions that we have ever seen. Policy assessment relies on real time monitoring of the economy using seasonally adjusted time series. This poses a new problem for analysts: to deseasonalize time series under the effects of strongly changing trends and seasonal patterns in real time. We compare different seasonal adjustment strategies based on parametric as well as non-parametric procedures. Firstly, using basic structural time series models we simulate data where we know their generating processes for the trend, cycle, seasonal and irregular components. In this way, we are able to have the counterfactual time series or what would have happened had COVID-19 not occurred. Afterwards, we contaminate the last 12 observations in different ways to reproduce the heterogeneity in the behaviour of time series during COVID-19 and check the performance of several alternatives for deseasonalizing.

LOW COMPLEXITY RECURRENT NEURAL NETWORKS FOR EDGE COMPUTING

This paper is dedicated to the development of recurrent neural networks in order to supplement the AI-based devices like microcontrollers and other mist computing systems. Due to the insignificant computational power of the edge deviсes the aim of the study is to design and analyze low complexity sequence models for a basic sensory time series forecasting on an example of univariate indoor temperature data. The description of data preparation and transformation followed by the models configuration via different architectures like simple LSTM and GRU is provided. To calculate an optimal set of hyper-parameters for the multiple neural network architectures a genetic algorithm has been implemented. The results of numerical experiments conducted for each model configuration consisting of both unidirectional and bidirectional cell connections are discussed. In addition to these studies the scheme of deploying the developed low-complexity models on STM32 microcontroller joined with the high-performance hub is proposed. Key words : edge computing, recurrent neural networks, time series, genetic algorithm.

Baseline Accuracies of Forecasting COVID-19 Cases in Russian Regions on a Year in Retrospect Using Basic Statistical and Machine Learning Methods

The large amount of data accumulated so far on the dynamics of the COVID-19 outbreak has allowed assessing the accuracy of forecasting methods in retrospect. This work compares several basic time series analysis methods, including machine learning methods, for forecasting the number of confirmed cases for some days ahead. Year-long data for all regions of Russia has been used from the Yandex DataLens platform. As a result, accuracy estimates for these basic methods have been obtained for Russian regions and Russia as a whole, in dependence on the forecasting horizon. The best basic models for forecasting for 14 days are exponential smoothing and ARIMA, with an error of 11–19% by the MAPE metric for the latest part of the course of the epidemic. The accuracies obtained can be considered as baselines for more complex prospective models.

Application of GARCH model in the price inflation of foodstuff in West Java

The risk and uncertainty faced by consumers and producers due to fluctuations in food prices in recent years have continued to increase. One of them is the fluctuation in the price of foodstuffs which are basic human needs. The process of increasing prices in general and continuously related to the market mechanism which can be caused by various factors is called inflation. Inflation can be measured by calculating the change in the percentage rate of change in a price index. In this study, the index used is the Consumer Price Index (CPI), which is an index that measures the average price of certain goods purchased by consumers. The food price index tends to have high volatility, this can be seen from the sharp increase and decrease in the price of food commodities in Indonesia. Data that has high volatility will be very risky if you use the basic time series method which has stationary data and costly variance, for that we need a method to deal with data that has non-constant and changing variance, or is called heteroscedasticity. Therefore in this study using the GARCH model application to analyze data on foodstuff price inflation in West Java. The results showed that the best GARCH model for foodstuff price inflation data is the ARCH (1) model, which means that the variance of foodstuff price inflation in period t is determined by a constant (1.565) and the remaining square in the previous period with a proportion of 74.41. %.

Testing Mean Stability of Heteroskedastic Time Series

Time series models are often fitted to the data without preliminary checks for stability of the mean and variance, conditions that may not hold in much economic and financial data, particularly over long periods. Ignoring such shifts may result in fitting models with spurious dynamics that lead to unsupported and controversial conclusions about time dependence, causality, and the effects of unanticipated shocks. In spite of what may seem as obvious differences between a time series of independent variates with changing variance and a stationary conditionally heteroskedastic (GARCH) process, such processes may be hard to distinguish in applied work using basic time series diagnostic tools. We develop and study some practical and easily implemented statistical procedures to test the mean and variance stability of uncorrelated and serially dependent time series. Application of the new methods to analyze the volatility properties of stock market returns leads to some unexpected surprising findings concerning the advantages of modeling time varying changes in unconditional variance.

Pengembangan E-Authentic Asessment Berbasis PBL untuk Meningkatkan Kompetensi Mahasiswa dalam Pembelajaran Jaringan Komputer

This study aims to develop authentic online assessment instruments based on problem based learning. Authentic assessment instruments will be accessed online in computer network learning to measure student competencies effectively and efficiently. Type of research is research and development and used refer to the 4 D model (define, design, develop, disseminate). The analysis technique uses the Aiken'V test, and validity uses expert testing and Focus Group Discussion (FGD). The practicality test is carried out by applying the product to students in the form of a product practicality questionnaire and to test the effectiveness using basic time series design. Validity test results show the RPS and instruments developed are valid with Aiken's V values ​​of 0.708 and 0.716. Practicality test results on the practicality test by lecturers and students on the RPS were 93% and authentic assessment 91%. Effectiveness test results with time series analysis were 38% with the effective category. The implication of the study is that there is a positive effect on learning outcomes, the value of students completeness increases and the gain score test is 0.38 in the medium category. The findings of this study have resulted in e-authentic assessment instruments with valid, practical and effective categories.