Comparative Time Series Analysis Research Articles

Autonomous Curtailment Control in Distributed Generation Planning

While enabling prominent growth of renewable-based distributed generation (DG) and matching energy harvesting, nonfirm connection policy requires accompanying active and reactive power dispatch control. Centralized control often implies comprehensive communication infrastructure and notable capital investment. Hence, DG planning models tailored for decentralized management of thermal and voltage constraints are found wanting. This paper presents a DG planning methodology, which successfully incorporates a new decentralized active power curtailment control and also includes decentralized voltage control. It optimizes autonomous control settings for each generator individually, such that the net energy export from the distribution to transmission level, per unit of installed capacity, is maximized. Comparative time series analysis validates the methodology.

Benchmarking workforce productivity in the creative industries

The competitive dynamics of many industries have changed considerably over the past decade and perhaps none more so than in publishing. There is no doubt that this industry continues to undergo structural changes that compel businesses to adapt and transform themselves to the demands and disruption caused by innovative new media technologies. But how have the workforce responded to these dynamic changes and what has been the impact on industry performance? This paper presents the findings from a comparative time-series analysis (1997-2013) of the UK Publishing Industry and compares it to other industries categorised within the UK Creative Industries. In doing so, it extends our understanding of inter-industry workforce performance and provide an insight into how the UK Publishing Industry has adapted to digital media over two decades change and turbulence.

Comparative analysis of time series of marine litter surveyed on beaches and the seafloor in the southeastern North Sea

The comparative analysis of marine litter in different marine compartments has rarely been attempted. In this study, long-term time series of marine litter abundance on the seafloor and on the coast, both from the southeastern North Sea, were analyzed for temporal trends and correlations. On four beach sections of 100 m length, mean abundances of total beach litter collected four times a year from 2002 to 2008 varied between 105 and 435 items. Mean densities of total inorganic litter on the seafloor amounted to 10.6 ± 9.7 kg km−2 in the offshore region (2001–2010) and 13.7 ± 12.6 kg km−2 in the Wadden Sea (1998–2007), respectively. In the offshore region, there was no significant long-term trend, while in the Wadden Sea, densities of marine litter declined significantly. Correlations between time series were weak, indicating different sources and transport processes responsible for compositions of beach litter and litter on the seafloor. Decreases in inputs from fisheries and substantial export due to resuspension are discussed as reasons for the decrease in litter on the seafloor in the Wadden Sea.

Suicides in Russia and Belarus: A Comparative Analysis of Trends

Background: The Slavic countries of the former Soviet Union (fSU) Russia and Belarus retain one of the highest suicide rates in the world, despite a gradual decline over the past decade. Aim: The present study aims to analyze whether population drinking is able to explain the dramatic fluctuations in suicide mortality in Russia and Belarus from the late Soviet to post-Soviet period. Method: Trends in suicide rates and alcohol sales per capita from 1970 to 2014 in Russia and Belarus were analyzed employing an ARIMA analysis. Results: Alcohol sales is a statistically significant associated with suicide rates in both countries, implying that a 1-l increase in per capita alcohol sales is associated with an increase in the suicide rates of 5.0% in Russia and of 6.1% in Belarus. Conclusion: This is the first comparative time-series analysis of alcohol sales and suicide rates in Russia and Belarus, which highlighted close temporal association between suicide rates and population drinking in both countries.

Alcoholic Psychoses in Russia and Belarus: A Comparative Analysis of Trends

Background: The alcoholic psychoses incidence rate is a reliable indicator of alcohol-related problems at the population level since there is a strong relationship between alcoholic psychoses incidence rate and alcohol consumption per capita. Aims: To estimate the aggregate level effect of alcohol on the psychoses incidence rate in the former Soviet Republics Russia and Belarus. Method: Trends in alcoholic psychoses incidence rate and alcohol sales per capita from 1970 to 2013 in Russia and Belarus were analyzed employing an ARIMA analysis. Results: Alcohol sales is a statistically significant associated with alcoholic psychoses incidence rate in both countries, implying that a 1-l increase in per capita alcohol sales is associated with an increase in the alcoholic psychoses incidence rate of 18.6% in Russia and of 18.8% in Belarus. Conclusion: This is the first comparative time-series analysis of alcohol sales and alcoholic psychoses incidence rate in Russia and Belarus, which highlighted close temporal association between alcoholic psychoses rate and population drinking in both countries.

Bayesian interpolation of unequally spaced time series

A comparative analysis of time series is not feasible if the observation times are different. Not even a simple dispersion diagram is possible. In this article we propose a Gaussian process model to interpolate an unequally spaced time series and produce predictions for equally spaced observation times. The dependence between two observations is assumed a function of the time differences. The novelty of the proposal relies on parametrizing the correlation function in terms of Weibull and Log-logistic survival functions. We further allow the correlation to be positive or negative. Inference on the model is made under a Bayesian approach and interpolation is done via the posterior predictive conditional distributions given the closest \(m\) observed times. Performance of the model is illustrated via a simulation study as well as with real data sets of temperature and CO\(_2\) observed over 800,000 years before the present.

Comparative time-series analysis of MeV electron data by Ulysses and Pioneer 10/11 in the Jovian magnetosphere

Abstract. The dynamics of the Jovian magnetosphere is dominated by the planet's fast rotation with a period of ~ 10 h. Within the magnetosphere, this periodicity can in particular be seen in the temporal variation of the spectral index of MeV electrons: every ~ 10 h the counting rates show a maximum (minimum), while the spectral index shows a minimum (maximum) known as the Jovian "clock" mechanism. In this study we re-analyse Ulysses and Pioneer 10/11 data and show that another periodic modulation in the MeV electrons can be identified, manifested by local maxima of the spectral index and local minima of the counting rates. For Ulysses, this modulation can be observed throughout the magnetosphere near the magnetic equator, suggesting an azimuthal asymmetric distribution of MeV electrons near the current sheet. This modulation is found to trail the "clock" mechanism by ~ 3.25 h. The Pioneer 10 data, however, only show occasional evidence of the presence of these local maxima while there is no evidence of this modulation in the Pioneer 11 data. A comparison of the times of observed minor peaks and Ulysses' distance from the current sheet using a simple rigid disc model as well as the model of Khurana and Schwarzl (2005) is performed.

Highly comparative time-series analysis: the empirical structure of time series and their methods

The process of collecting and organizing sets of observations represents a common theme throughout the history of science. However, despite the ubiquity of scientists measuring, recording and analysing the dynamics of different processes, an extensive organization of scientific time-series data and analysis methods has never been performed. Addressing this, annotated collections of over 35 000 real-world and model-generated time series, and over 9000 time-series analysis algorithms are analysed in this work. We introduce reduced representations of both time series, in terms of their properties measured by diverse scientific methods, and of time-series analysis methods, in terms of their behaviour on empirical time series, and use them to organize these interdisciplinary resources. This new approach to comparing across diverse scientific data and methods allows us to organize time-series datasets automatically according to their properties, retrieve alternatives to particular analysis methods developed in other scientific disciplines and automate the selection of useful methods for time-series classification and regression tasks. The broad scientific utility of these tools is demonstrated on datasets of electroencephalograms, self-affine time series, heartbeat intervals, speech signals and others, in each case contributing novel analysis techniques to the existing literature. Highly comparative techniques that compare across an interdisciplinary literature can thus be used to guide more focused research in time-series analysis for applications across the scientific disciplines.

Comparative Time Series Analysis of Perceptual Responses to Electroacoustic Music

this study investigates the relationship between acoustic patterns in contemporary electroacoustic compositions, and listeners' real-time perceptions of their structure and affective content. Thirty-two participants varying in musical expertise (nonmusicians, classical musicians, expert computer musicians) continuously rated the affect (arousal and valence) and structure (change in sound) they perceived in four compositions of approximately three minutes duration. Time series analyses tested the hypotheses that sound intensity influences listener perceptions of structure and arousal, and spectral flatness influences perceptions of structure and valence. Results suggest that intensity strongly influences perceived change in sound, and to a lesser extent listener perceptions of arousal. Spectral flatness measures were only weakly related to listener perceptions, and valence was not strongly shaped by either acoustic measure. Differences in response by composition and musical expertise suggest that, particularly with respect to the perception of valence, individual experience (familiarity and liking), and meaningful sound associations mediate perception.

The occurrence of floods and the role of climate variations from 1880 in Calabria (Southern Italy)

Abstract. In this paper, we present a methodological approach based on a comparative analysis of floods that occurred in a wide region over a long period and the climatic data characterising the same period, focusing on the climate trend. The method simplifies the comparative analysis of several time series by defining some indexes (e.g. the monthly, bi-monthly, and ... m-monthly indexes of precipitation, temperature, wet days and precipitation intensity and the monthly flood number) that can be used to study phenomena such as floods that are characterised by spatial and temporal variability. The analysis was used to investigate the potential effect of climate variation on the damaging floods trend. The approach was tested for the Calabria region (Italy) using historical flood and climatic data from 1880 to 2007. The results showed that the number of floods was correlated with the monthly indexes of precipitation, wet days, and daily precipitation intensity. The following trends were recognised: decreasing precipitation and wet days, almost constant precipitation intensity, increasing temperature, and linearly increasing floods. A second-order polynomial trend analysis showed a slight decrease in floods since the seventies, which might be explained by the favourable climatic conditions during the period and/or the effect of increasing awareness of flood vulnerability.

Occurrence of landslide events and the role of climate in the twentieth century in Calabria, southern Italy

Abstract A methodological approach based on analysing landslides that occurred over a long period and climatic data characterizing that period is presented. The method investigates whether there are any effects of climate on landslide triggering. The approach has been tested in Calabria (Italy). Both landslide and climatic data have been obtained from available databases that have been expanded. Landslide data came from historical archives and newspapers, whereas the climatic analysis is based on daily and monthly series of rainfall and temperature. The method simplifies the comparative analysis of several time series by defining some indices (the monthly, bi-monthly … m -monthly indices of precipitation, temperature, wet days and precipitation, and the monthly landslide number) that can be used to study phenomena, such as landslides, that are characterized by spatial and temporal variability. For Calabria, the number of landslides is correlated to monthly precipitation, wet days and precipitation intensity. Thus, landslide occurrence could be roughly forecast using these climatic data. Despite the favourable climatic trend, landslides are not decreasing because the recent utilization of landslide-prone areas increases the vulnerability.

Economics of apple-storage I: Comparative time series analysis of apple producer prices in Germany and Hungary

Based on standard econometric methods the article analyses the time series of fresh apple producer prices in domestic markets of Germany and Hungary. In Germany, as a consequence of high storage capacities the quantity offered in different parts of the season is relatively stable, that’s why only a rather limited price increasing can be detected. In Hungary, as a consequence of the limited storage capacity this fluctuation is much more important. The modern methods of time series analysis (ARIMA models, stepwise regression) can be efficiently applied for forecasting of price movements.

Seasonal regularities of the observed warming in Siberia

This paper discusses the results of a comparative analysis of a multiyear time series for average seasonal and annual surface temperatures in Siberia. To analyze seasonal temperature fluctuations, we use the data on average seasonal temperatures which excludes the influence of higher-frequency (cyclonic and daily) fluctuations. We have obtained an empirical correlation between the seasonal fluctuations and average seasonal surface temperatures for two climatic zones (Irkutsk and Tobol’sk). For both zones, the seasonal fluctuations amplitude is, with a high determination coefficient (over 0.9), in linear correlation with the average seasonal winter temperature. We point out that it is promising to model and forecast the observed climatic change on the basis of empirical regularities.

Use of Alternative Pollutant Temporal Metrics in Time Series Analyses of Ambient Air Pollution and Cardiorespiratory Emergency Department Visits

P-065 Introduction: In studying the temporal relationship between ambient air pollution levels and health outcomes, choosing an appropriate daily pollution metric may not be straightforward. The use of a given time metric in an epidemiologic analysis may be justified by biological plausibility, spatial representativeness, human time-activity patterns or data availability. In order to describe the variability of epidemiologic results attributable to the use of different pollutant metrics, we conducted a comparative time-series analysis using multiple daily pollutant metrics for the same pollutant. Methods: We obtained hourly measurements of ambient particulate and gaseous air pollutants (i.e., NO2, O3, PM10, ultrafine PM, PM2.5, PM2.5 sulfate, PM2.5 total carbon, PM2.5 elemental carbon, CO, and SO2) from existing air monitoring networks for the time period 1993–2004. Data on emergency department visits for cardiorespiratory outcomes were available from 41 hospitals in the study area. For an initial analysis, we investigated NO2, a primary pollutant exhibiting diurnal and spatial variation, and cardiovascular disease (CVD), which were associated in our previous epidemiologic investigation using a 1-hour maximum metric. The following temporal metrics were created: 1) a 1-hour daily maximum; 2) a 24-hour average; 3) a “rush hour” period, defined as the average pollution between 7-9am and 4-6pm; 4) a day-time (8am-6pm) period; and 5) a night-time (12am-5am) period. In addition, NO2 levels in each hour of the day were examined separately. Using Poisson generalized linear models, we examined the association of these NO2 metrics with daily counts of cardiovascular disease visits. Results: Ongoing analyses of multiple pollutants will be presented; here we present the NO2 metric comparison. Correlations between the metrics were:TableRisk ratios for an interquartile range increases in metric-specific NO2 were 1.022 (χ2=28.2) for 1-hour maximum, 1.021 (χ2=23.6) for 24-hour average, 1.018 (χ2=16.2) for rush hour, 1.018 (χ2=16.2) for night-time, and 1.013 (χ2=8.5) for day-time. Individual hour NO2 risk ratios for a 20ppb increase ranged from 0.99 (χ2=0.48) at 3pm to 1.022 (χ2=25.32) at 5am. No associations were observed with individual hours between 1pm and 5pm, hours with typically low NO2 concentrations. Discussion: Effect estimates varied according to the choice of time metric used to characterize daily NO2 pollution levels. Implications of this heterogeneity and results from similar analyses of other pollutants and outcomes will be discussed.

Large-Scale Comparative Analysis of Pertussis Population Dynamics: Periodicity, Synchrony, and Impact of Vaccination

Pertussis is a worldwide infectious disease which persists despite massive vaccination campaigns that have gone on for several decades. To obtain an overall view of pertussis dynamics and the impact of vaccination, the authors performed, using the wavelet method, a comparative analysis of pertussis time series in 12 countries to detect and quantify periodicity and synchrony between them. Results showed a clear 3- to 4-year cycle in all countries, but the main finding was that this periodicity was transient. No global pattern in the effect of vaccination on pertussis dynamics was observed, but some spatial synchrony between countries was detected. This large-scale comparative analysis of pertussis dynamics sheds light on the complexity of the multiple interactions involved in global pertussis spatial dynamic patterns. It suggests a need to perform a global survey of human infectious diseases over the long term, which would permit better assessment of the risk of disease outbreaks in the future.

Multiscaling comparative analysis of time series and geophysical phenomena

AbstractDifferent methods are used to determine the scaling exponents associated with a time series describing a complex dynamical process, such as those observed in geophysical systems. Many of these methods are based on the numerical evaluation of the variance of a diffusion process whose step increments are generated by the data. An alternative method focuses on the direct evaluation of the scaling coefficient of the Shannon entropy of the same diffusion distribution. The combined use of these methods can efficiently distinguish between fractal Gaussian and Lévy‐walk time series and help to discern between alternative underling complex dynamics. © 2005 Wiley Periodicals, Inc. Complexity 10: 51–56, 2005

The use of international data on fuel sales and vehicle fleet for the estimation of yearly national traffic volumes

International comparisons are frequently used to assess the road safety performance of a country or to monitor its development over time, at a national or regional level. In most instances either accident or injury risks have to be considered, which results in the need for a quantitative estimation of the amount of travel, namely traffic volumes. Only in special cases may risks be indirectly compared using methods of induced exposure. The required data on traffic volumes is not always readily available. This is especially true if old data or data from several countries for a long period are to be used, as in comparative time series analysis. Several simplifying assumptions can be adopted to bypass this lack of data but often the estimates resulting from these simple assumptions are not entirely satisfactory. Between 1996 and 1998 the European Union (EU) sponsored, within the scope of the 4th Research Framework Programme, a COST Action (COST 329) to analyse and establish a common methodological framework for description and analysis of safety developments—past and future—that could also represent the effect of interventions. Researchers from 14 European countries participated in this action. One activity carried out was the development of a method for estimating the traffic volume of countries where such data is not available, using data on vehicle fleet and fuel sales in the studied country and mathematical models fitted to existing data (from other countries) on fuel consumption, vehicle fleet and traffic volume. In the European context, this method is especially useful for some Eastern and Southern European countries. In this paper the mentioned method is described, the results of its application to Portugal are presented and ways to use it in other countries to estimate their traffic volumes for the period 1980–2000 are described.

Multiscaling comparative analysis of time series and a discussion on "earthquake conversations" in California.

Time series are characterized by complex memory and/or distribution patterns. In this Letter we show that stochastic models characterized by different statistics may equally well reproduce some pattern of a time series. In particular, we discuss the difference between Lévy-walk and fractal Gaussian intermittent signals and show that the adoption of complementary scaling analysis techniques may be useful to distinguish the two cases. Finally, we apply this methodology to the earthquake occurrences in California and suggest the possibility that earthquake occurrences are described by a colored ("long-range correlated") generalized Poisson model.

Measuring system normality

A comparative analysis of transaction time-series is made, for light to moderately loaded hosts, motivated by the problem of anomaly detection in computers. Criteria for measuring the statistical state of hosts are examined. Applying a scaling transformation to the measured data, it is found that the distribution of fluctuations about the mean is closely approximated by a steady-state, maximum-entropy distribution, modulated by a periodic variation. The shape of the distribution, under these conditions, depends on the dimensionless ratio of the daily/weekly periodicity and the correlation length of the data. These values are persistent or even invariant. We investigate the limits of these conclusions, and how they might be applied in anomaly detection.

Alcohol consumption and suicide mortality by age among Finnish men, 1950-1991.

An earlier comparative time series analysis of the connection between per capita alcohol consumption and suicide mortality reported, unexpectedly, a non-significant effect of alcohol on the suicide rate of Finnish men. However, the suicide rate of Finnish men is heterogeneous in regard to age groups. There is also indication that the connection between alcohol and suicides is stronger in younger than in older age groups. Time series data on per capita alcohol consumption and age-specific suicide mortality in 1950-91 were used to scrutinize the relationship. The results indicate that the suicide rate of Finnish men in age groups 15-34 years and 35-49 years is associated with per capita alcohol consumption, while no connection could be established between the suicide rates of men aged 50-69 years and 70+ years and either per capita alcohol consumption or a lagged consumption measure. Real income, divorce rate and unemployment rate were regarded as possible confounding variables.