Frequency Time Series Research Articles

Seasonality in High Frequency Time Series

Time series observed at higher frequencies than monthly frequency display complex seasonal patterns that result from the combination of multiple seasonal patterns (with annual, monthly, weekly and daily periodicities) and varying periods, due to the irregularity of the calendar. Seasonality in high frequency data is modelled from two main perspectives: the stochastic harmonic approach, based on the Fourier representation of a periodic function, and the time-domain random effects approach. An encompassing representation illustrates the conditions under which they are equivalent. Three major challenges are considered: the first deals with modelling the effect of moving festivals, holidays and other breaks due to the calendar. Secondly, robust estimation and filtering methods are needed to tackle the level of outlier contamination, which is typically high, due to the lower level of temporal aggregation and the raw nature of the data. Finally, model selection strategies play an important role, as the number of harmonic or random components that are needed to account for the complexity of seasonality can be very large.

Increase in the number of explosive low‒level cyclones around King George Island in the last three decades.

This paper documents an increase in the number of observed explosive cyclones (EC) at King George Island, South Shetland Islands, Antarctica, over the 1989‒2020 period. In ECs at 60o latitudes the surface atmospheric pressure drops ≥24 hPA in 24 hours. The annual EC frequency time series shows a significant positive trend of ~2.7 cyclones/decade, with a break in 2003 and average numbers of 7.3 and 11.8 events before and after that break, respectively. The increase follows closely earlier documented global sea surface temperature (SST) anomaly trends for the 1981‒2018 period, partially attributed to global warming and to the Pacific Decadal Oscillation (PDO). Connections between EC frequency and SST might occur through variations in SST in the southeastern Pacific and southwestern Atlantic, with anomalous cold conditions favoring an increase in ECs. We also found close relations between the number of ECs with simultaneous occurrences of PDO and Atlantic multidecadal oscillation in opposite phases, so that after 2003 they were in the cold and warm phases, respectively, and vice-versa before 2003. Both low-frequency modes seem to modulate the number of ECs. As per the authors knowledge these results have not been discussed before and may help climate modeling studies and weather forecasts.

Effects from the June 10, 2021 solar eclipse in the ionosphere over Kharkiv: results from ionosonde measurements

Background. Ionospheric effects from solar eclipses (SEs) have been studied for over a century. The studies intensified in the 1960–1970, when rockets, satellites, ionosonde networks, and incoherent scatter radars became actively involved in the investigations. The following basic features of SEs displayed in the ionosphere have been established: a decrease in the electron density and the electron, ion, and neutral temperatures, changes in the dynamics of the ionosphere, and the generation of acoustic and atmospheric gravity waves. The recurring regular features are observed together with the features pertaining to each individual SE. Therefore, the study of the effects of each new solar eclipse is an urgent task. The purpose of this paper is to present the results of analysis of the ionogram measurements acquired by the digisonde located at the Radiophysical Observatory, V. N. Karazin Kharkiv National University. Techniques and Methodology. The observations of the state of the ionosphere were made using the digisonde specifically developed by the V. N. Karazin Kharkiv National University and located at the V. N. Karazin Kharkiv National University Radiophysical Observatory (49°38' N, 36°20' E). The data processing included the following. First, the time series of critical F2-layer frequencies, foF2, of the minimum frequency, fmin, observed on an ionogram, and of the virtual height were plotted. Second, the trends and the differences between the initial series and the trends were calculated. Third, given the foF2(t) values, the electron densities N(t) and their increments ΔN(t) were calculated, and fourth, the systems spectral analysis of ΔN(t) dependences was conducted to determine the spectral content. Results. A maximum decrease of 3.7–3.8% in the F2 peak electron density has been detected. This value agrees very well with the calculated value of 3.5–3.8%. The time delay between the minimum in the electron density observed and the maximum value of the phase of the solar eclipse was determined to be 12.5 min, which is in good agreement with the calculated value of 12.8 min. The time delay suggests that the linear loss coefficient was 1.3´10–3 s–1 and the production rate was (3.8–3.9)×108 m–3∙s–1. The SE was associated with the generation of quasi-periodic oscillations in the electron density and the F2 layer virtual height, within the period range of 10–15 min, and with amplitudes of 1.7–4% and 9.1–11.4%, respectively. The fluctuations in the minimum frequency observed on the ionograms showed a factor of 3–4 times decrease in the course of the SE. In addition, a decrease in the averaged values was observed to occur from 3 to 2.85 MHz (by 5%), which suggested that the radio wave attenuation by absorption decreased due to a decrease in the lower-ionosphere electron density below 100 km altitude. This decrease was estimated to be about 2%, while the decrease determined from the observations gave 2.2%. Conclusions. The partial solar eclipse with a very small value of the maximum magnitude (0.112) and obscuration of 4.4% acted to produce a few effects observed in the ionosphere with the ionosonde.

An analytical framework based on the recency, frequency, and monetary model and time series clustering techniques for dynamic segmentation

Nowadays, banks use data mining and business intelligence tools and techniques to analyze their customers’ behavior. Customer segmentation is a widely adopted analytical tool to identify distinct groups of customers. Most studies have adopted a static segmentation approach, leading to missing important trends and patterns. In this study, we propose a framework that represents each customer behavior as a time-series sequence of the Recency, Frequency, and Monetary variables, and then exploits time-series clustering algorithms to carry out customer segmentation. The framework consists of state-of-the-art clustering algorithms, including hierarchical, spectral, and k-shape algorithms to divide customers into homogeneous groups and implement point of sale devices transaction data of grocery and appliance retailers. We divide customers into four segments, analyze the behavioral trends, and provide marketing suggestions for each segment. Our results show that, in terms of the computed validity indices, the best clustering model for grocery retailers is reached using hierarchical clustering with the Complexity-Invariant Distance measure, and for appliance retailers, the best segmentation is achieved by applying the spectral clustering with the Complexity-Invariant Distance measure.

Temporal and Spatial Patterns of Harmful Algae Affecting Scottish Shellfish Aquaculture

Consistent patterns of Harmful Algal Bloom (HAB) events are not evident across the scientific literature, suggesting that local or regional variability is likely to be important in modulating any overall trend. This study summarizes Scotland-wide temporal and spatial patterns in a robust 15-year high temporal frequency time series (2006–2020) of the incidence of HABs and shellfish biotoxins in blue Mussels (Mytilus edulis), collected as part of the Food Standards Scotland (FSS) regulatory monitoring program. The relationship between the countrywide annual incidence of HAB events and biotoxins with environmental variables was also explored. Temporal patterns exhibited interannual variability, with no year-on-year increase, nor any correlation between annual occurrences. Within years, there was a summer increase in bloom frequency, peaking in July for Dinophysis spp. and Pseudo-nitzschia spp., and a plateau from May to July for Alexandrium spp. Temporal-spatial patterns were analyzed with multivariate statistics on data from monitoring sites aggregated monthly into 50-km grid cells, using Principal Component Analysis (PCA) and cluster K-means analysis. PCA analyses showed correlation between areas with similar temporal dynamics, identifying seasonality as one of the main elements of HAB variability with temporal-spatial patterns being explained by the first and second principal components. Similar patterns among regions in timing and magnitude of blooms were evaluated using K-means clusters. The analysis confirmed that the highest risk from HABs generally occurred during summer, but demonstrated that areas that respond in a similar manner (high or low risk) are not always geographically close. For example, the occurrence of the most prevalent HAB genus, Dinophysis spp., is similar countrywide, but there is a regional trend in risk level with “very-high” and “high” clusters located primarily on the southwest coast, the islands of the central and northern west coast and the Shetland Islands. “Early” and “late” blooms were also associated with certain areas and level of risk. Overall, high risk areas mainly face in a southwest direction, whilst low risk locations face a south or southeast direction. We found relatively few countrywide relationships between environmental variables and HABs, confirming the need for regional analysis to support HAB early warning.

The capabilities of analyzing the seismo-electromagnetic satellite CSES-01 data for monitoring of seismic activity of the Northern Tien Shan

The relationship between space weather and earthquakes occurring in seismically active areas with the seismo-electromagnetic satellite CSES-01 data are observed. The study assumes the presence of possible ionospheric precursors of strong earthquakes. The sets of measuring instruments of the satellite and data are considered. Subsequently, to prove the non-random nature of the space weather effects, the results of ground-based geomagnetic or electromagnetic observations should be used, for example, magnetotelluric soundings (MTS). Stationary and regime points of the MTS network of the Research Station RAS in Bishkek (RS RAS) are suitable for these purposes in the Northern Tien Shan. The MTS data are presented in the form of hourly frequency-time series with system rotation from 0° to 180°. To obtain information on earthquakes, it is proposed to use data from the NEIC or ISC seismic catalog, since they aggregate data from regional catalogs, including the KNET seismological network maintained by the RS RAS. It is concluded that it is necessary to select the CSES-01 trajectories and times of flight over the territory of Kyrgyzstan and download the electric field detector (EFD) data. The results of comparing satellite and ground-based observations can be used in future technologies for short-term earthquake prediction.

Using implementation of artificial intelligence in estimating the exchange rate in the foreign exchange market

The Foreign Exchange currency market is considered the most liquid market in the world due to the amount of money that is traded every day. Researchers and investors are interested in having a system that allows predicting the direction of the exchange rate of a pair of currencies in order to create a strategy that allows them to profit from the market. Since the movement of the exchange rate between currencies generates a high frequency time series, recurrent neural networks have gained strength in the prediction of these markets, since they have the characteristic of generating models of non-linear systems. In this document, the study of the prediction of the direction of the exchange rate of different currencies was carried out using the recurrent auto-regressive neural network with exogenous inputs, which has the characteristics to obtain a predictive model of a time series. The tests were carried out in the Euro / Dollar pair in the time frame of one hour, demonstrating the potential of this neural network architecture. These techniques are products of the biological and physical sciences in industrial processes.

Research Trends in College English Education in Korea －A Topic Analysis Using LDA Topic Modeling

This study investigated the research trends of college English education in Korea from 2001 to 2020. The data was collected using a Biblio data collector and a total of 313 papers were analyzed. For research purposes, the data were analyzed using frequency analysis, LDA (Latent Dirichlet Allocation), and time series analysis. The summary of the findings is as follows: In the first instance, the number of research papers regarding college English education has increased significantly in quantity for 20 years. Secondly, in analyzing the topics of the chosen papers, a total of 10 topics in college English education were found. The topics were “curriculum and level-differentiated programs (T1)”, “learners’ affective factors (T2)”, “assesment and learning strategies (T3)”, “teachers’ factors (T4)”, “English vocabulary, grammar and writing (T5)”, “English for specific purposes (T6)”, “teaching and learning methods (T7)”, “web-based learning (T8)”, “learner-centered education (T9)”, and “textbook analysis etc. (T10).” Among these topics, the three that were identified as topics increasing in popularity were “learners’ affective factors (T2)”, “English for specific purposes (T6)”, and “learner-centered education (T9).” The topics increasing in popularity shared one key characteristic: the topics were related to learners’ factors such as the learners’ motivation, the learners’ goals, and the learners’ activities in class. This study is meaningful in that it collected a wide range of data related to college English education in Korea and produced more reliable results by using big data-based LDA topic modeling techniques.

Multivariate temporal data analysis ‐ a review

ABSTRACTThe information technology revolution, especially with the adoption of the Internet of Things, longitudinal data in many domains become more available and accessible for secondary analysis. Such data provide meaningful opportunities to understand process in many domains along time, but also challenges. A main challenge is the heterogeneity of the temporal variables due to the different types of data, whether a measurement or an event, and type of samplings: fixed or irregular. Other variables can be also events that may or not have duration. In this review, we discuss the various types of temporal data, and the various relevant analysis methods. Starting with fixed frequency variables, with forecasting and time series methods, and proceeding with sequential data, and sequential patterns mining, and time intervals mining for events having various time duration. Also the use of various deep learning based architectures for temporal data is discussed. The challenge of heterogeneous multivariate temporal data analysis and discuss various options to deal with it, focusing on an increasingly used option of transforming the data into symbolic time intervals through temporal abstraction and the use of time intervals related patterns discovery for temporal knowledge discovery, clustering, classification prediction, and more. Finally, we discuss the overview of the field, and areas in which more studies and contributions are needed.This article is categorized under: Algorithmic Development > Spatial and Temporal Data Mining

텍스트 마이닝을 활용한 영어 쓰기교육 연구동향 분석

This study investigated research trends of English writing education in Korea from 2001 to 2020. The data was collected using a Biblio data collector and a total of 1,495 papers were analyzed. For research purposes, they were analyzed using frequency analysis, LDA(Latent Dirichlet allocation), and time series analysis. The summary of the findings is as follows. In the first instance, the number of research papers on English writing has increased significantly in quantity for 20 years and a total of 126 academic journals published research papers on English writing. Secondly, the topics of the papers were analyzed. There were a total of 25 topics in English writing education. The topics were

Frequency stability with converter-connected resources delivering fast frequency control

Converter-connected resources, such as electric vehicles and stationary batteries, are being recognized by grid operators as contributors to frequency stability. Their fast-response potential makes them ideal candidates to provide fast frequency control.In this paper we discuss and present evidence on what kind of dynamics a power system experiences when using electric vehicles to provide frequency control. Initially, we present a method to assess a system balance by deriving so called system residuals, given historical frequency time series and components characteristics. The system residuals consist of the non-controllable, and not explicitly measured, part of generation and load. Afterwards, the Danish island of Bornholm, where vehicles are used on a commercial basis to provide primary frequency control, is used as the test-bed for the investigation. The analysis is supported by validating the described components and system models against real measurements from the islanded power system.The key outcome is that the overall response time needs to be within 1 second. For the considered system scenarios, the share of reserve provided by electric vehicles should be up to 30% to improve the system stability. We finally generalize the results by highlighting key factors needed to replicate the findings in other power systems.

An exploratory data quality analysis of time series physiologic signals using a large-scale intensive care unit database.

Physiological data, such as heart rate and blood pressure, are critical to clinical decision-making in the intensive care unit (ICU). Vital signs data, which are available from electronic health records, can be used to diagnose and predict important clinical outcomes; While there have been some reports on the data quality of nurse-verified vital sign data, little has been reported on the data quality of higher frequency time-series vital signs acquired in ICUs, that would enable such predictive modeling. In this study, we assessed the data quality issues, defined as the completeness, accuracy, and timeliness, of minute-by-minute time series vital signs data within the MIMIC-III data set, captured from 16009 patient-ICU stays and corresponding to 9410 unique adult patients. We measured data quality of four time-series vital signs data streams in the MIMIC-III data set: heart rate (HR), respiratory rate (RR), blood oxygen saturation (SpO2), and arterial blood pressure (ABP). Approximately, 30% of patient-ICU stays did not have at least 1 min of data during the time-frame of the ICU stay for HR, RR, and SpO2. The percentage of patient-ICU stays that did not have at least 1 min of ABP data was ∼56%. We observed ∼80% coverage of the total duration of the ICU stay for HR, RR, and SpO2. Finally, only 12.5%%, 9.9%, 7.5%, and 4.4% of ICU lengths of stay had ≥ 99% data available for HR, RR, SpO2, and ABP, respectively, that would meet the three data quality requirements we looked into in this study. Our findings on data completeness, accuracy, and timeliness have important implications for data scientists and informatics researchers who use time series vital signs data to develop predictive models of ICU outcomes.

A mutual information based R-vine copula strategy to estimate VaR in high frequency stock market data.

In this paper, we explore mutual information based stock networks to build regular vine copula structure on high frequency log returns of stocks and use it for the estimation of Value at Risk (VaR) of a portfolio of stocks. Our model is a data driven model that learns from a high frequency time series data of log returns of top 50 stocks listed on the National Stock Exchange (NSE) in India for the year 2014. The Ljung-Box test revealed the presence of Autocorrelation as well as Heteroscedasticity in the underlying time series data. Analysing the goodness of fit of a number of variants of the GARCH model on each working day of the year 2014, that is, 229 days in all, it was observed that ARMA(1,1)-EGARCH(1,1) demonstrated the best fit. The joint probability distribution of the portfolio is computed by constructed an R-Vine copula structure on the data with the mutual information guided minimum spanning tree as the key building block. The joint PDF is then fed into the Monte-Carlo simulation procedure to compute the VaR. If we replace the mutual information by the Kendall's Tau in the construction of the R-Vine copula structure, the resulting VaR estimations were found to be inferior suggesting the presence of non-linear relationships among stock returns.

Exploring the role of conventional energy consumption on environmental quality in Brazil: Evidence from cointegration and conditional causality

The study explores the nexus between energy, export, import, population and economic growth on environmental quality in Brazil. Hence the long-run equilibrium association and dynamic causality amongst environmental quality—proxy with CO2 emission—energy consumption, trade policy, and population growth in Brazil is empirically estimated. Annual frequency time-series data from 1971 to 2016 is employed within the ARDL bounds testing methodological framework. The conditional Granger causality procedure within the VECM is followed to examine dynamic short-term and long-run causations in the estimated model. Thus, a stable long-run relationship is empirically established in the estimated model. Hence, within the CO2 emission energy-augmented model—via the channel of real GDP per capita, real per capita exports/imports, and population growth—CO2 emissions converge to its long-run equilibrium by an average speed of 37.47% on an annually basis. Additionally, a 1% increase in energy consumption increases CO2 emissions by 1.259%. Similarly, an increase in GDP growth and export worsens environmental quality by increasing CO2 emissions by 0.033% and 2.202% respectively. The result of the impulse responses and variance decomposition in response to exogenous shocks in the model collaborate the findings of the ARDL model. Shock to energy use, real GDP per capita, real exports per capita, real imports per capita, and population growth causes changes in environmental quality—both in the short-run and long-run period—with significant implication for environmental conservation in Brazil. The short-run causality supports the imperative of energy for economic growth. Thus, suggests caution in following conservation policy so as not to jeopardize real economic growth, whilst contemplating environmental sustainability in Brazil.

Robust estimation for Binomial conditionally nonlinear autoregressive time series based on multivariate conditional frequencies

A topical problem of robust statistical estimation of parameters for binomial conditionally nonlinear autoregressive (BiCNAR) time series under innovation outliers is considered. This problem is solved by means of s-order Markov properties for observed time series and probabilistic properties of multivariate conditional frequencies of the future state under its s-prehistory. The new robust statistical estimator ζˆ called frequencies-based estimator (FBE) is constructed for the BiCNAR parameters under innovation outliers with arbitrary discrete probability distribution having some fixed known expectation. Under mild regularity conditions the constructed FBE is shown to have the robustness properties: consistency and asymptotic normality with obtained asymptotic covariance matrix. FBE also has computational advantages: an explicit form and a fast recursive re-estimation algorithm for extension of the model. Asymptotic risk functional and its minimum are evaluated using Fisher information matrix for the considered model. Sensitivity analysis of the statistical estimator ζ̃ for the BiCNAR parameters, that is constructed for the hypothetical model without outliers, is carried out for the situation with innovation outliers: ζ̃ is shown to be inconsistent in this situation, its bias and the instability coefficient are evaluated and analyzed. The robust estimator ζˆ has a free parameter — weight matrix H. The optimal weight matrix H∗ is found by minimization of the asymptotic risk w.r.t. H. Statistical estimator for H∗ based on the observed time series is constructed. Results of multiple computer experiments on simulated and real data illustrate the theory.

Prediction error of Johansen cointegration residuals for structural health monitoring

A novel method for structural health monitoring under environmental and operational variations (EOV) is proposed based on the prediction errors of the Johansen cointegartion (CI) residuals using a Recurrent Neural Network (RNN). The first four natural frequency time series of the structure, identified from vibration measurements over a period of time, are used to this end. The Variational Mode Decomposition (VMD) algorithm is first used for denoising and removing seasonal patterns in the frequency signals. The first modes of the decomposition results corresponding to all frequency signals are then used to obtain Johansen CI residuals. Next, a portion of the obtained signals form VMD decomposition along with the same portion of the Johansen CI residuals are used respectively as training features and targets to train a RNN. The trained RNN is then used to predict the future CI residuals from the remaining portion of the features. The error of the prediction result is used as damage sensitive feature. The proposed method has been successfully tested on a long-term monitoring problem of a numerical example (spring-mass system), a short-term monitoring problem regarding an experimental example (wooden bridge), and a long-term monitoring of an experimental example (the Z24 bridge). The results demonstrate the capability of the proposed method in monitoring structures for damage even when the Johansen algorithm fails to identify a linear CI relationship among the frequency signals.

Internal migration and mobile communication patterns among pairs with strong ties

Using large-scale call detail records of anonymised mobile phone service subscribers with demographic and location information, we investigate how a long-distance residential move within the country affects the mobile communication patterns between an ego who moved and a frequently called alter who did not move. By using clustering methods in analysing the call frequency time series, we find that such ego-alter pairs are grouped into two clusters, those with the call frequency increasing and those with the call frequency decreasing after the move of the ego. This indicates that such residential moves are correlated with a change in the communication pattern soon after moving. We find that the pre-move calling behaviour is a relevant predictor for the post-move calling behaviour. While demographic and location information can help in predicting whether the call frequency will rise or decay, they are not relevant in predicting the actual call frequency volume. We also note that at four months after the move, most of these close pairs maintain contact, even if the call frequency is decreased.

Tracking COVID-19 using online search

Previous research has demonstrated that various properties of infectious diseases can be inferred from online search behaviour. In this work we use time series of online search query frequencies to gain insights about the prevalence of COVID-19 in multiple countries. We first develop unsupervised modelling techniques based on associated symptom categories identified by the United Kingdom’s National Health Service and Public Health England. We then attempt to minimise an expected bias in these signals caused by public interest—as opposed to infections—using the proportion of news media coverage devoted to COVID-19 as a proxy indicator. Our analysis indicates that models based on online searches precede the reported confirmed cases and deaths by 16.7 (10.2–23.2) and 22.1 (17.4–26.9) days, respectively. We also investigate transfer learning techniques for mapping supervised models from countries where the spread of the disease has progressed extensively to countries that are in earlier phases of their respective epidemic curves. Furthermore, we compare time series of online search activity against confirmed COVID-19 cases or deaths jointly across multiple countries, uncovering interesting querying patterns, including the finding that rarer symptoms are better predictors than common ones. Finally, we show that web searches improve the short-term forecasting accuracy of autoregressive models for COVID-19 deaths. Our work provides evidence that online search data can be used to develop complementary public health surveillance methods to help inform the COVID-19 response in conjunction with more established approaches.

Technical Note: Improved partial wavelet coherency for understanding scale-specific and localized bivariate relationships in geosciences

Abstract. Bivariate wavelet coherency is a measure of correlation between two variables in the location–scale (spatial data) or time–frequency (time series) domain. It is particularly suited to geoscience, where relationships between multiple variables differ with locations (times) and/or scales (frequencies) because of the various processes involved. However, it is well-known that bivariate relationships can be misleading when both variables are dependent on other variables. Partial wavelet coherency (PWC) has been proposed to detect scale-specific and localized bivariate relationships by excluding the effects of other variables but is limited to one excluding variable and provides no phase information. We aim to develop a new PWC method that can deal with multiple excluding variables and provide phase information. Both stationary and non-stationary artificial datasets with the response variable being the sum of five cosine waves at 256 locations are used to test the method. The new method was also applied to a free water evaporation dataset. Our results verified the advantages of the new method in capturing phase information and dealing with multiple excluding variables. Where there is one excluding variable, the new PWC implementation produces higher and more accurate PWC values than the previously published PWC implementation that mistakenly considered bivariate real coherence rather than bivariate complex coherence. We suggest the PWC method is used to untangle scale-specific and localized bivariate relationships after removing the effects of other variables in geosciences. The PWC implementations were coded with Matlab and are freely accessible (https://figshare.com/s/bc97956f43fe5734c784, last access: 14 January 2021).

Clean energy development in the United States amidst augmented socioeconomic aspects and country-specific policies

The drive toward the attainment of sustainable environment globally through clean energy development or energy efficiency is not more desirable than in the 21st century, thus the existential policy moderations of economic, trade and security mechanisms. On this premise, and foremost in the literature, the current study examined the country-specific (for the United States) and the driving impacts of economic policy uncertainty, trade policy and national security on the development of cleaner energy sources by using quarterly frequency time series data for period 1990:Q1-2018:Q2. By employing economic expansion as additional factor, the study implemented the Autoregressive Distributed Lag Bounds Testing approach to reveal interesting results: (1) there is a significant evidence that economic expansion, economic policy uncertainty (EU), trade policy (TP), and national security (NS) exhibits long term properties in common, (2) the increase in economic expansion and NS effectiveness significantly yields more cleaner energy development, and (3) a more tightened TP and high EU are statistically significant and detrimental to the development of clean energy. The Granger causality evidence substantiates the role economic expansion, TP, EU and national security in renewable energy development. Generally, the study posits cleaner and energy efficiency policy directive for policymakers in the United States and other countries of interest from the framework of climate action and sustainable development.