Frequency Time Series Research Articles

Glacier retreat changes diurnal variation intensity and frequency of hydrologic variables in Alpine and Andean streams

The impacts of glacier mass and area loss are assumed to affect stream hydrology and related ecosystems. We applied wavelet analysis to high frequency time series of discharge, conductivity, turbidity, air and water temperature in mountain streams influenced by a gradient of diverse glacier catchment coverages from 0 to 28%. All parameters were recorded during 200–250 day period in 24 mountain streams in Switzerland and Ecuador. We interpreted the range of glacier coverage as a proxy of glacier melt effect linked to modification in glacier cover based on a space-for-time substitution approach. We used the diurnal variation power and frequency as hydrological parameters to quantify the intensity and occurrence of significant diurnal variations in the different mountain catchments. Our results show that the reduction in glacier cover would reduce the intensity and occurrence of diurnal variations in discharge, turbidity and conductivity, and the intensity of diurnal variations in water temperature. Discharge and conductivity were affected in a similar way, while the impact on turbidity was less significant. Furthermore, due to their different climatic regimes, the diurnal discharge in the tropical Andes was more impacted than in the Alps, indicating that the impacts of global change on glacier streams would be stronger in the tropical Andes than in the Alps.

DC Series Arc Fault Detection Algorithm for Distributed Energy Resources Using Arc Fault Impedance Modeling

Arc fault detection is important technology to guarantee the safety of power systems and is therefore essential for producing practical power systems for real-world applications. However, fuses and arc fault detection devices (AFDD) struggle to detect series arc faults in DC systems, because the series arc fault induces small current variation between the normal and abnormal conditions. In addition, switching noise from the grid-connected inverter makes detecting arc fault conditions even more difficult. This paper proposes arc fault detection algorithm based on the relative comparison of current variability in terms of frequency spectrum and time series. The operational principle of the proposed algorithm is analyzed to detect the arc fault condition. In addition, the investigation of arc fault impedance using the small-signal modeling can obtain the resonant frequency of arc fault condition at low frequency range. From the impedance model, the frequency analysis range can be designed to avoid the switching noise of inverter. The performance of proposed arc fault detection algorithm is verified with a 3.8 kW grid-connected PV system and arc fault generator.

Machine Learning Panel Data Regressions with an Application to Nowcasting Price Earnings Ratios

This paper introduces structured machine learning regressions for nowcasting and forecasting with panel data consisting of series sampled at different frequencies. Motivated by the empirical problem of predicting corporate earnings for a large cross-section of firms with macroeconomic, financial, and news time series sampled at different frequencies, we focus on the sparse-group LASSO regularization. This type of regularization can take advantage of the mixed frequency time series panel data structures and we find that it empirically outperforms the unstructured machine learning methods. We obtain oracle inequalities for the pooled and fixed effects sparse-group LASSO panel data estimators recognizing that financial and economic data can have fat tails. To that end, we leverage on a new Fuk-Nagaev concentration inequality for panel data consisting of heavy-tailed $\tau$-mixing processes. Our empirical results show the superior performance of introduced machine learning panel data regressions over analysts' predictions, forecast combinations, firm-specific time series regression models, and unstructured elastic net.

Predictability of Power Grid Frequency

The power grid frequency is the central observable in power system control, as it measures the balance of electrical supply and demand. A reliable frequency forecast can facilitate rapid control actions and may thus greatly improve power system stability. Here, we develop a weighted-nearest-neighbor (WNN) predictor to investigate how predictable the frequency trajectories are. Our forecasts for up to one hour are more precise than averaged daily profiles and could increase the efficiency of frequency control actions. Furthermore, we gain an increased understanding of the specific properties of different synchronous areas by interpreting the optimal prediction parameters (number of nearest neighbors, the prediction horizon, etc.) in terms of the physical system. Finally, prediction errors indicate the occurrence of exceptional external perturbations. Overall, we provide a diagnostics tool and an accurate predictor of the power grid frequency time series, allowing better understanding of the underlying dynamics.

COVID-19 Impact on Cryptocurrencies: Evidence from a Wavelet-Based Hurst Exponent

Cryptocurrency history begins in 2008 as a means of payment proposal. However, cryptocurrencies evolved into complex, high yield speculative assets. Contrary to traditional financial instruments, they are not (mostly) traded in organized, law-abiding venues, but on online platforms, where anonymity reigns. This paper examines the long term memory in return and volatility, using high frequency time series of eleven important coins. Our study covers the pre-COVID-19 and the subsequent pandemia period. We use a recently developed method, based on the wavelet transform, which provides more robust estimators of the Hurst exponent. We detect that, during the peak of COVID-19 pandemic (around March 2020), the long memory of returns was only mildly affected. However, volatility suffered a temporary impact in its long range correlation structure. Our results could be of interest for both academics and practitioners.

Spatial analysis of spring dust storms hazard in Iran

In this research, monthly data on the frequency of dust storms of 44 synoptic stations of Iran during a 50-year period from 1968 to 2017 and methods of Ward’s hierarchical cluster analysis, linear regression, Mann–Kendall test, and kriging geostatistic were used. After calculating the frequency of occurrence of dust storms and its related statistical characteristics, suitable probability distribution function was fitted with dust storm data and then, using cluster analysis, the classification of different regions of the country into homogeneous groups was done and by using the kriging method, the zoning of dust storms was performed. After examining the descriptive statistics of spring dust storms in Iran, it was found that the probability distribution of dust storm data occurrence is proportional to the probability distribution of the normal 3-parameter log normal. According to the results of cluster analysis, different regions of the country were clustered into five homogeneous regions comprising of (1) southeast regions, (2) southern and central regions, (3) western regions, (4) eastern and middle regions, and (5) the northern regions. In investigating the trend of time variations of dust storms, it was revealed that the frequency time series of dust storms in Iran had an increasing trend and these results were confirmed by Mann–Kendall statistics. After zoning, it was found that the main focus of this phenomenon was concentrated in the southeast and southern parts of the country.

A study on trade-offs between spatial resolution and temporal sampling density for wheat yield estimation using both thermal and calendar time

Within-season forecasting of crop yields is of great economic, geo-strategic and humanitarian interest. Satellite Earth Observation now constitutes a valuable and innovative way to provide spatio-temporal information to assist such yield forecasts. This study explores different configurations of remote sensing time series to estimate of winter wheat yield using either spatially finer but temporally sparser time series (5daily at 100 m spatial resolution) or spatially coarser but denser (300 m and 1 km at daily frequency) time series. Furthermore, we hypothesised that better yield estimations could be made using thermal time, which is closer to the crop physiological development. Time series of NDVI from the PROBA-V instrument, which has delivered images at a spatial resolution of 100 m, 300 m and 1 km since 2013, were extracted for 39 fields for field and 56 fields for regional level analysis across Northern France during the growing season 2014-2015. An asymmetric double sigmoid model was fitted on the NDVI series of the central pixel of the field. The fitted model was subsequently integrated either over thermal time or over calendar time, using different baseline NDVI thresholds to mark the start and end of the cropping season. These integrated values were used as a predictor for yield using a simple linear regression and yield observations at field level. The dependency of this relationship on the spatial pixel purity was analysed for the 100 m, 300 m and 1 km spatial resolution. At field level, depending on the spatial resolution and the NDVI threshold, the adjusted R² ranged from 0.20 to 0.74; jackknifed – leave-one-field-out cross validation – RMSE ranged from 0.6 to 1.07 t/ha and MAE ranged between 0.46 and 0.90 t/ha for thermal time analysis. The best results for yield estimation (adjusted R² = 0.74, RMSE =0.6 t/ha and MAE =0.46 t/ha) were obtained from the integration over thermal time of 100 m pixel resolution using a baseline NDVI threshold of 0.2 and without any selection based on pixel purity. The field scale yield estimation was aggregated to the regional scale using 56 fields. At the regional level, there was a difference of 0.0012 t/ha between thermal and calendar time for average yield estimations. The standard error of mean results showed that the error was larger for a higher spatial resolution with no pixel purity and smaller when purity increased. These results suggest that, for winter wheat, a finer spatial resolution rather than a higher revisit frequency and an increasing pixel purity enable more accurate yield estimations when integrated over thermal time at the field scale and at the regional scale only if higher pixel purity levels are considered. This method can be extended to larger regions, other crops, and other regions in the world, although site and crop-specific adjustments will have to include other threshold temperatures to reflect the boundaries of phenological activity. In general, however, this methodological approach should be applicable to yield estimation at the parcel and regional scales across the world.

Analysis of the probability distribution of time series of word frequencies

Knowledge of probability distribution of frequency of words and phrases is important in solving many practical problems such as estimation of semantic similarity between words, detection of semantic changes and others. It is assumed by many researchers that frequencies of words obey the Poisson law. However, there is much evidence that the Poisson distribution describes empirical data unsatisfactory. The analysis of the probability law in this case is greatly complicated by the fact that series of frequencies in most cases are non-stationary. This paper discusses the distribution law of time series of word frequencies based on the Google Books Ngram corpus data. It is shown that the correlation between the first moments of the frequencies differs from that expected in the assumption of the Poisson distribution. In particular, anomalously high values of frequency dispersion are observed for words with high frequency. To check the significance of deviations from the Poisson law, statistical modeling of frequency series was performed.

Processing, analysis and interpretation of time-frequency series for magnetotelluric monitoring

This work presents the results of processing and interpretation of data of magnetotelluric monitoring obtained in the territory of the Bishkek geodynamic station (Northern Tien Shan). A new approach to the analysis of magnetotelluric monitoring data is proposed. The correlation of the obtained time-frequency series of variations of electromagnetic parameters with the moon-solar tidal effects, deformations of the Earth’s crust, variations of apparent resistance and the distribution of seismicity was performed. A concept, linking the change in the stress-strain state of the medium with the redistribution of fluids between fracture systems, which causes variations in electrical resistance, was confirmed.

The Delphi technique in forecasting– A 42-year bibliographic analysis (1975–2017)

Since the infancy of the Delphi Technique for collecting and aggregating expert insight, this methodological tool has been discussed, adapted and applied in over 2,600 published scholarly papers to date. This paper mines the major citation indexing services to analyze five dimensions of these data: primary contribution (methodological or applied), field and subfield, length (in pages), year, and journal/conference. Interpreted visual analytics of these five dimensions (both individually and in combination) provide researchers, practitioners and editors with clear insights about whether the Delphi technique is still as prominently used, discussed, and written about in the academic literature as it was twenty years ago and the related trends that might inform predictions of its future use. Among these insights, a simple time series of frequencies of Delphi publications by year immediately shows that academic acceptance of Delphi as a research tool is not only well established, but it has been growing in popularity and range of research domains for two decades predicting unprecedented levels of use in the years to come.

Dry Spells and Extreme Precipitation are The Main Trigger of Landslides in Central Europe

Landslides are frequently triggered by extreme meteorological events which has led to concern and debate about their activity in a future greenhouse climate. It is also hypothesized that dry spells preceding triggering rainfall may increase slope predisposition to sliding, especially in the case of clay-rich soils. Here we combined dendrogeomorphic time series of landslides and climatic records to test the possible role of dry spells and extreme downpours on process activity in the Outer Western Carpathians (Central Europe). To this end, we tested time series of past frequencies and return periods of landslide reactivations at the regional scale with a Generalized Linear Mixed (GLM) model to explore linkages between landslide occurrences and triggering climate variables. Results show that landslide reactivations are concentrated during years in which spring and summer precipitation sums were significantly higher than usual, and that triggering mechanisms vary between different types of landslides (i.e. complex, shallow or flow-like). The GLM model also points to the susceptibility of landslide bodies to the combined occurrence of long, dry spells followed by large precipitation. Such situations are likely to increase in frequency in the future as climate models predict an enhancement of heatwaves and dry spells in future summers, that would be interrupted by less frequent, yet more intense storms, especially also in mountain regions.

Online dynamic group-buying community analysis based on high frequency time series simulation

Group-buying often fails even when there are satisfactory quantities as not enough consumers join in the required time, which can waste seller, purchaser, and platform operator time resources; therefore, the group buying features require further research. Over a 3 weeks period, around 700 million click-stream records from 1,061,770 users from a stable and continuous time series were allocated to groups of 5 min frequency, and a hybrid neural network model developed to simulate group-buying behavior in four experiments, from which it was found that adding to the cart and adding as a favorite were significant group-buying behavior features, and shopping depth was the main demographic feature, but age was not. Compared with previous ambiguous online consumer feature conclusions on gender, the results revealed that the commodity feature was the main determinant for gender feature significance. The college student feature was found to be a pseudo feature, and should connect with other fixed effects such as low income or education level. This paper is the first to construct an online dynamic group-buying community, which is a new type of social network and could provide a new perspective for social commerce research. A big data neural network-based method for examining group-buying community behavior over time is proposed that can offer novel insights to online vendors for the development of targeted marketing campaigns.

A new methodology for customer behavior analysis using time series clustering

Purpose The purpose of this paper is to propose a new methodology that handles the issue of the dynamic behavior of customers over time. Design/methodology/approach A new methodology is presented based on time series clustering to extract dominant behavioral patterns of customers over time. This methodology is implemented using bank customers’ transactions data which are in the form of time series data. The data include the recency (R), frequency (F) and monetary (M) attributes of businesses that are using the point-of-sale (POS) data of a bank. This data were obtained from the data analysis department of the bank. Findings After carrying out an empirical study on the acquired transaction data of 2,531 business customers that are using POS devices of the bank, the dominant trends of behavior are discovered using the proposed methodology. The obtained trends were analyzed from the marketing viewpoint. Based on the analysis of the monetary attribute, customers were divided into four main segments, including high-value growing customers, middle-value growing customers, prone to churn and churners. For each resulted group of customers with a distinctive trend, effective and practical marketing recommendations were devised to improve the bank relationship with that group. The prone-to-churn segment contains most of the customers; therefore, the bank should conduct interesting promotions to retain this segment. Practical implications The discovered trends of customer behavior and proposed marketing recommendations can be helpful for banks in devising segment-specific marketing strategies as they illustrate the dynamic behavior of customers over time. The obtained trends are visualized so that they can be easily interpreted and used by banks. This paper contributes to the literature on customer relationship management (CRM) as the proposed methodology can be effectively applied to different businesses to reveal trends in customer behavior. Originality/value In the current business condition, customer behavior is changing continually over time and customers are churning due to the reduced switching costs. Therefore, choosing an effective customer segmentation methodology which can consider the dynamic behaviors of customers is essential for every business. This paper proposes a new methodology to capture customer dynamic behavior using time series clustering on time-ordered data. This is an improvement over previous studies, in which static segmentation approaches have often been adopted. To the best of the authors’ knowledge, this is the first study that combines the recency, frequency, and monetary model and time series clustering to reveal trends in customer behavior.

Diurnal to interannual variability of sea surface pCO2 and its controls in a turbid tidal-driven nearshore system in the vicinity of the East China Sea based on buoy observations

We examined the diurnal to seasonal dynamics of the sea surface partial pressure of carbon dioxide (pCO2) in a subtropical nearshore estuarine system, Hangzhou Bay, adjacent to the Changjiang Estuary in the vicinity of the East China Sea, based on data collected between July 30, 2010 to September 20, 2011 by a surface buoy equipped with an autonomous pCO2 system along with hydrological and other chemical sensors. The study site (122.37° E, 30.55° N) is influenced by the river plumes of both the Changjiang and Qiantang River and is characterized by strong tidal circulation and highly turbid waters. The amplitude of pCO2 changes increased from winter to summer over both diurnal and spring-neap tidal cycle timescales. The average surface water pCO2 was slightly undersaturated with respect to the atmosphere in winter (382 ± 18 μatm), but supersaturated in spring (500 ± 56 μatm) and summer (687 ± 110 μatm). Overall the study site was a source of atmospheric CO2 with an average sea to air flux of 14 ± 9 mmol C m−2 d−1 from January to October 2011. We revealed factors controlling the pCO2 dynamics at different timescales. Over seasonal timescales, temperature and estuarine mixing dominated the seawater pCO2 variability. Over spring-neap tidal timescales in winter and spring, the major drivers were similarly water mass mixing and temperature. However, in summer, biological activity and air-sea exchange became the two principal factors controlling the variations in surface seawater pCO2. Our mass balance models further suggested that biological processes impacted surface pCO2 differently during different tidal phases. Respiration was revealed to promote the increase in pCO2 during spring tide in August, but in neap tides of the same month biological production was evident and resulted in the drawdown of pCO2. This is because photosynthesis was generally limited by light in summer at the study site due to high turbidity, except during neap tides when turbidity was dramatically drawn down, triggering high biological productivity. At the diurnal timescale, sea surface pCO2 was primarily controlled by tidal mixing, except during neap tides in summer when sea surface pCO2 was greatly influenced by biological metabolism. This study also revealed significant inter-summer differences between 2010 and 2011, showing lower sea surface pCO2 in August 2010 as compared to August 2011, which was likely due to the enhanced biological uptake as a result of the relatively low turbidity caused by weak tidal currents and enhanced river flow in August 2010. Our study highlights a highly dynamic system primarily driven by tidal mixing, which not only modulates water mass mixing but also affects turbidity, which subsequently controls biological production. These processes led to a synergy of CO2 dynamics in a tidally driven and highly turbid nearshore system, where high frequency time-series observations are essential to reveal the complex controls of CO2 dynamics.

Delivering combined droop and inertial response from wind plant for power system frequency stability

As wind penetration level into power systems increases, provision of frequency support from wind plant is anticipated to become a requirement in grid codes. In addition to the virtual inertial response, two different approaches to delivering droop response from wind plant are presented and compared in this paper. The combination of droop and inertial response from wind plant has been assessed for a frequency time series. Simulation results show that the proposed approaches to delivering combined droop and inertial response can provide superior performance to either inertial or droop response alone.

CDS İLE DÖVİZ KURLARI ARASINDAKİ UZUN DÖNEMLİ İLİŞKİ: TÜRKİYE ÖRNEĞİ

The main objective of this study is to analyze the long-term relationship between exchange rate and credit default swaps (CDS) based on high frequency time series as representative of Turkey's risk premium. In general, the long-term relationship describes a long period of time in the literature. However, although the time periods used in the testing of financial and economic hypotheses have taken a long period of time, the use of new techniques is needed due to the high frequencies. In the analysis of long-term relationships with high-frequency time series, the transition and the continuity of the shocks should be considered together. Therefore, in this study, partial integration method considering these two features was used as an analysis tool. The most important feature of the analysis is that it gives information about the long term relationship based on these properties. This information is a random process based on the tendency to revert to the mean in the period covered and is the result of the applied test. Hence, the relationship between variables, namely exchange rates and Turkey's CDS, are analyzed by using non-linear causality tests. Thus, it is also analyzed whether the effects such as jump and break on these variables change over time. Policy recommendations are made for Turkey based on the empirical findings to contribute to the relevant literature.

Exploring primary school teachers’ technological pedagogical content knowledge (TPACK) in online collaborative discourse: An epistemic network analysis

AbstractThe contextual influences on technological pedagogical content knowledge (TPACK) enactment and the method of TPACK assessment remain to be important research topics. Discourse data of 81 teachers in an online professional learning community were collected and analyzed based on the framework of TPACK. Frequency distribution and time series characteristics of teachers’ knowledge domains were analyzed. In addition, epistemic network analysis was used to compare the epistemic network characteristics of teachers in the higher‐score and the lower‐score groups, different age groups, and post and reply groups. Results showed that teachers’ knowledge domains enacted in the context of online discourse were mainly pedagogical content knowledge and general pedagogical knowledge. The teachers in the higher‐score group had a rich, organized and flexible knowledge structure of TPACK. Younger teachers had more connections between pedagogical knowledge and pedagogical content knowledge, while senior teachers had more connections between technological knowledge and pedagogical knowledge. The teachers in the reply group had more connections among the different categories of knowledge as compared to the post group. Finally, implications, limitations and future research were discussed.

The forecasting evaluation of the high-order mixed frequency time series model to the marine industry

The forecasting evaluation of the high-order mixed frequency time series model to the marine industry

A framework for automated anomaly detection in high frequency water-quality data from in situ sensors

Monitoring the water quality of rivers is increasingly conducted using automated in situ sensors, enabling timelier identification of unexpected values or trends. However, the data are confounded by anomalies caused by technical issues, for which the volume and velocity of data preclude manual detection. We present a framework for automated anomaly detection in high-frequency water-quality data from in situ sensors, using turbidity, conductivity and river level data collected from rivers flowing into the Great Barrier Reef. After identifying end-user needs and defining anomalies, we ranked anomaly importance and selected suitable detection methods. High priority anomalies included sudden isolated spikes and level shifts, most of which were classified correctly by regression-based methods such as autoregressive integrated moving average models. However, incorporation of multiple water-quality variables as covariates reduced performance due to complex relationships among variables. Classifications of drift and periods of anomalously low or high variability were more often correct when we applied mitigation, which replaces anomalous measurements with forecasts for further forecasting, but this inflated false positive rates. Feature-based methods also performed well on high priority anomalies and were similarly less proficient at detecting lower priority anomalies, resulting in high false negative rates. Unlike regression-based methods, however, all feature-based methods produced low false positive rates and have the benefit of not requiring training or optimization. Rule-based methods successfully detected a subset of lower priority anomalies, specifically impossible values and missing observations. We therefore suggest that a combination of methods will provide optimal performance in terms of correct anomaly detection, whilst minimizing false detection rates. Furthermore, our framework emphasizes the importance of communication between end-users and anomaly detection developers for optimal outcomes with respect to both detection performance and end-user application. To this end, our framework has high transferability to other types of high frequency time-series data and anomaly detection applications.

Maximum Likelihood Estimation of Fitness Components in Experimental Evolution.

Estimating fitness differences between allelic variants is a central goal of experimental evolution. Current methods for inferring such differences from allele frequency time series typically assume that the effects of selection can be described by a fixed selection coefficient. However, fitness is an aggregate of several components including mating success, fecundity, and viability. Distinguishing between these components could be critical in many scenarios. Here, we develop a flexible maximum likelihood framework that can disentangle different components of fitness from genotype frequency data, and estimate them individually in males and females. As a proof-of-principle, we apply our method to experimentally evolved cage populations of Drosophila melanogaster, in which we tracked the relative frequencies of a loss-of-function and wild-type allele of yellow This X-linked gene produces a recessive yellow phenotype when disrupted and is involved in male courtship ability. We find that the fitness costs of the yellow phenotype take the form of substantially reduced mating preference of wild-type females for yellow males, together with a modest reduction in the viability of yellow males and females. Our framework should be generally applicable to situations where it is important to quantify fitness components of specific genetic variants, including quantitative characterization of the population dynamics of CRISPR gene drives.