Application Of Time Series Analysis Research Articles

Time series characterization of simulated microtubule dynamics in the nerve growth cone.

The process of neurite outgrowth is critically dependent on proper microtubule assembly. However, characterizing the dynamics of microtubule assembly and their quantitative relationship to neurite outgrowth is a difficult task. The difficulty can be reduced by using time series analysis which has broad application in characterizing the dynamics of stochastic, or "noisy," behaviors. Here we apply time series analysis to quantitatively compare simulated microtubule assembly and neurite outgrowth in vitro. Microtubule length life histories were simulated assuming constant growth and shrinkage rates coupled with random selection of growth and shrinkage times, a formulation based on the dynamic instability model of microtubule assembly. Net length displacements of simulated microtubules were calculated at discrete, evenly spaced times, and the resulting time series were characterized by both spectral and autocorrelation analysis. Depending on the sampling rate and the dynamic parameters, simulated microtubules exhibited significant autocorrelation and periodicity. To make a comparison to neurite outgrowth, we characterized the dynamic behavior of simulated microtubule populations and found it was not significantly different from that of single microtubules. The net displacements of rat superior cervical ganglion neurite tips were measured and characterized using time series methods. Their behavior was consistent with the microtubule dynamics for appropriate simulation parameters and sampling rates. Our results show that time series analysis can provide a useful tool for quantitative characterization of microtubule dynamics and neurite outgrowth and for assessing the relationship between them.

Estimation of reference change limits using patient data.

Two approaches for deriving reference change limits from patient data are described. In the direct method, hospital database information is used for the selection of appropriate reference groups. If database information is not sufficient or reliable enough, but still most of the source data can be considered as health-related, an indirect method can be applied in the calculation of rough estimates for reference change limits. A computer program developed by us, GraphROC for Windows, includes both methods for the estimation of change limits from patient data. Time between specimen collections should be included as one classifying factor in the selection of source data. When only one previous result is available for comparison, change limits based on the reference sample group form the only available guide for clinical interpretation. However, when several previous results are available and the within-subject variances for the considered analyte are known to be heterogeneous between individuals, the clinical interpretation should rather be based on application of time series analysis.

Application of time series analysis in the clinical setting.

A growth curve time series model is characterized by a biological growth pattern with random deviations added to this basic pattern. Various types are reviewed including the homeostatic model which is a special and simple type characterized by zero growth. The limitations to the use of time series analysis in the clinical setting include lack of initial baseline values and difficulties in specifying in quantitative terms a relevant hypothesis alternative to the one postulated by the time series model. Some ongoing research addressing these problems is discussed.

Developing diagnostics for input-output systems: the effects of certain linear and nonlinear filters on the correlation integral

Abstract. Input-output systems are characterized by applying time series analysis techniques developed for autonomous systems to the input and the output time series separately and using the results as nonlinear statistics of the time series distributions. Two examples are presented using the correlation integral as a nonlinear statistic: the first one examines the change in the statistic when several sample input time series are passed through a nonlinear filter. The rectifier is chosen as the filter because it models, at first approximation, the effect of dayside magnetospheric reconnection to the interplanetary magnetic field and solar wind input. The changes in the correlation integral are used to characterize the filter response. A second example compares a linear filter of the rectified solar wind input to the observed auroral geomagnetic activity in terms of their correlation integrals. Implications for models of the solar wind-magneto- sphere coupling are discussed.

Measuring Economic Markets for Imported Crude Oil*

A previous paper by Weiner (1991) noted that many policy issues involving crude oil imports hinge on whether crude oil markets are unified or regionalized. Weiner observed that the literature on crude oil markets has paid little attention to the regionalization issue. However, a generalized literature addressing market delineation has been evolving for some time and recent advances in applied time series analysis have produced multivariate testing procedures which avoid most of the problems of the bivariate price correlation analyses previously employed in analyzing regionalization issues. This paper advances the work of Weiner by incorporating cointegration relationships into multivariate time series models and using these models to examine the extent of regionalization in the world market for crude oil imports. The empirical results reported here lend support to Adelman's characterization of the world oil market as “one great pool” (Adelman 1984).

Price Discovery in Asset Markets

Holbrook Working (1949) discovered that the percentage change in futures prices seemed to be largelyrandom. This led Paul Samuelson (1965) to develop the Efficient Market Hypothesis (EMH) which claims that the current spot and futures1 prices fully reflect all relevant information. Furthermore, because the future flow of information cannot be anticipated, price changes will not be serially correlated. These papers linked the notion of randomness of price changes to informational efficiency. From that point on, a major part of the empirical studies of asset markets has been the application of time series analysis to asset prices, in order to evaluate whether the price changes are random and whether futures prices reflect all available information. As the statistical tests became more sophisticated, the number of empirical studies increased and the results became more contradictory and difficult to interpret. An economic theorist can only be bemused by contemplating the empirical/econometric studies in the finance literature.

Two Reduced-Bias Autocorrelation Estimators: rF1 and rF2

Among the problems associated with the application of time-series analysis to typical psychological data are difficulties in parameter estimation. For example, estimates of autocorrelation coefficients are known to be biased in the small-sample case. Previous work by the present authors has shown that, in the case of conventional autocorrelation estimators of ρ1 the degree of bias is more severe than is predicted by formulas that are based on large-sample theory. Two new autocorrelation estimators, rF1 and rF2, were proposed; a Monte Carlo experiment was carried out to evaluate the properties of these statistics. The results demonstrate that both estimators provide major reduction of bias. The average absolute bias of rF2 is somewhat smaller than that of rF1 at all sample sizes, but both are far less biased than is the conventional estimator found in most time-series software. The reduction in bias comes at the price of an increase in error variance. A comparison of the properties of these estimators with those of other estimators suggested in 1991 shows advantages and disadvantages for each. It is recommended that the choice among autocorrelation estimators be based upon the nature of the application. The new estimator rF2 is especially appropriate when pooling estimates from several samples.

Evaluation applications of regression analysis with time-series data

Time-series analysis can be a powerful evaluation tool whose use was suggested as early as 1963 by Campbell and Stanley. A number of authors have used time-series analysis for evaluation purposes. For example, Gillings, Makuc, and Siegel (1981) used time-series analysis to assess the effect of regionalized perinatal care. Veney and Luckey (1983) used time-series analysis to evaluate the effect of changes in highway speed limit laws on highway accident deaths. Deutsch and Alt (1977) and Zimring (1975) used time-series analysis to assess the effect of gun control. Guerin and MacKinnon (1985) used it to examine child seat belt legislation. Valuable methodological contributions to the evaluation uses of timeseries analysis have been made by Mandell (1987) and by Sawyer (1987). Despite this work, which spans a period of nearly twenty years, time series analysis still receives little attention in the evaluation practice literature. The purpose of this paper is to describe the application of time-series analysis, and particularly the use of regression analysis for analyzing time series in a way that may make it more readily accessible to an evaluation practice audience. In particular, this paper will provide practical guidelines for applying regression analysis to time-series data of a type that may be readily available to decision makers in government agencies, health and social service agencies, ministries of health or social welfare in developing countries, or to other organizations that wish to better understand and evaluate program actions. The specific example used in this paper is drawn from the health sector and is an example from Sri Lanka concerned with tetanus toxoid immunization levels for expectant mothers and neonatal tetanus deaths. However, the techniques examined are broadly applicable to a variety of time-series analysis situations.’ The analysis methods discussed include linear and non-linear analysis with time as the independent variable, linear analysis with the level of tetanus toxoid immunizations as the independent variable, and a piecewise regression example that examines the interaction between time and both the start-up of a government immunization program and immunization levels. The paper also deals with problems of serial correlation of errors in time series and concludes with a discussion of interpretation and appropriate use of regression analysis in time-series situations.

Small Sample Properties of the Spearman Autocorrelation Estimator

The small sample properties of a variant of the Spearman rank correlation coefficient applied in the time-series context were investigated through Monte Carlo methods. The rank method ( r1S) has even greater bias than the highly biased conventional parametric procedure; a traditional test of H0: ρ1 = 0 based on ( r1S) yields unacceptable properties. Empirical small sample distributions associated with the rank coefficient differ markedly from the distributions predicted by asymptotic theory. It is concluded that neither rank nor conventional parametric estimators and hypothesis tests are appropriate for very small samples in applications of time-series analysis that have been recommended in the behavioral and social science literature.

The Application of Time Series Analysis to Assess and Control the Prevalence of Cirrhosis of Pigs Within Northern Ireland

The Application of Time Series Analysis to Assess and Control the Prevalence of Cirrhosis of Pigs Within Northern Ireland

Application of time-series analysis for the recognition of increases in urinary estrogens as markers for the beginning of the potentially fertile period

Time-series analysis was applied to the urinary total estrogen data from 142 ovulatory menstrual cycles to determine the first statistically significant increase as a marker for the beginning of the potentially fertile phase. Application of the Trigg's tracking signal to each cycle detected an increase in urinary total estrogens above the baseline in every case, with a cumulative probability of ≥95%. The distribution of first increase days ranged from 10 days before to 3.5 days before the presumed day of ovulation with a mean of 6.5 ± 1.4 days. The significant parameters in the calculation of the tracking signal are the smoothing constant (α) which is related to the number of baseline observations, the baseline mean for the cycle, and the variation of the baseline mean. The method allows a calculation of the tracking signal as the cycle unfolds and a statistical assessment can be given each day. The procedure is easily adaptable for computer calculation, and because it is applied to individual cycles avoids the use of population means with the loss of information inherently associated with the combination of data from many cycles. The Trigg's tracking signal is an appropriate method of analysis of menstrual cycle data and represents a satisfactory alternative to the more usual cumulative sum procedure. The distribution of the first increases constitutes a reference standard for urinary estrogen assays.

Padé approximation and its application in time series analysis

Since the appearance in 1970 of the book by Box and Jenkins, the use of the auto regressive-moving average (ARMA) model has become widespread in many fields for the analysis and prediction of time series data. However, one of the main criticisms of these ARMA models is that they are difficult to specify. In this paper we have proposed a new method based on the theory of Padé approximations for the estimation of parameters and specification of the order of ARMA ( p,q) models. Simulation results and results obtained from applying this method to real data sets substantiate our claim for using this method.

Modelling wastewater treatment plants through time series analysis

AbstractTime series analysis models are very useful in modelling dynamic systems in science and engineering applications. This class of models is in fact able to represent the dynamic features of physical systems that are subject to often uncontrollable inputs with random components. Wastewater treatment plants are examples of dynamic systems, with inputs (flow, organic loads, etc.) that vary stochastically within more or less wide ranges. The use of stochastic models allows a more detailed representation of the dynamic nature of these systems, while retaining the degree of information contained in most deterministic models.In this paper, time series analysis applications to wastewater treatment plant modelling are presented and discussed. Both univariate and multivariate stochastic processes are applied to sewage treatment plant data, and the results thus obtained are further analyzed and compared with those from “conventional” deterministic models.Specifically, the above mentioned models are analyzed with respect to possible application in the daily operation of sewage treatment plants, by virtue of their predictive capacities and relative ease of determination. Furthermore, these models present other attractive features, such as adaptiveness and the possibility of extracting useful information about the system from the analysis of their structure. Adaptiveness refers to the possibility of continuously improving the model's performance as new information about the system is collected by manual or automatic monitoring. Performance of the models herein identified, and possible applications of these models in real systems are discussed.

Application of Time-Series Analysis to Induced Gamma Ray Spectroscopy Logs From Two Cold Lake Heavy-Oil Observation Wells

Summary Induced gamma ray spectroscopy (IGRS) logs from two cyclic-steam-stimulation observation wells in Cold Lake were analyzed to determine the vertical resolution and repeatability of data derived from gamma rays of inelastic and capture neutron reac tions. Time-series analysis, a technique that uses the Fourier representation of the log data, was used to quantify the vertical resolution and the signal/noise characteristics of various IGRS log curves and to compute the coherence vs. spatial frequency of data collected on multiple IGRS passes. The coherence function ranges from 1.0 for perfect repeatability to 0.0 for incoherent noise. Because no real variations in the measured data were expected for the 12- to 24-hour data-collection period, any deviation of the coherence function from 1.0 is attributed to incoherent noise. Generally, coherence is high at low frequencies (large vertical scales) and low at high fre quencies (small vertical scales). By noting the frequency at which the coherence level decreases to the expected value of random noise, we can quantify the vertical resolution of a log curve. Data analysis from these wells indicates that both the vertical resolution and repeatability of individual capture and inelastic curves differ. We found that the H-yield and capture curves have the highest vertical resolution (≈0.3 m) and the best signal/noise ratios (FSN≈30:1). In contrast, the capture Ca and Si yields are of significantly lower quality (FSN≈2:1). Only a small difference exists between the vertical resolution of the inelastic C (1.0 m) and O (1.3 m) yields, but the FSN the O yield is only one-half that of the C yield. Fortunately, the vertical resolution and the repeatability of the C/O ratio, FCO, are determined primarily by the quality of the inelastic C data. Specific to our application of monitoring heavy-oil saturations during cyclic-steam stimulation in a relatively uniform, high-porosity unconsolidated sand, we developed a simplified FCO interpretation that does not rely on the elemental yields from capture data. This linear model permits us to assess the uncertainties in computed saturations from the statistical noise in the inelastic FCO data.

Estimating aquifer parameters from analysis of forced fluctuations in well level: An example from the Nubian Formation near Aswan, Egypt: 1. Hydrogeological background and large‐scale permeability estimates

This is the first of three papers concerning the extraction of aquifer parameters through the application of time series analysis to water level data from boreholes. The analysis is performed on data from six wells located in the Nubian desert near the shores of Lake Nasser, Egypt. These wells are ideally suited to this study since they are remote from most sources of hydrological noise and hence respond primarily to forcing functions that can be measured or modeled. For four of the wells, which were part of a monitoring system installed prior to impoundment of Lake Nasser, monthly (at least) water level measurements began in 1964. In 1985, two additional wells were drilled, and high precision piezometers that sampled data every 0.2 hour were installed in five of the wells. Thus data are available which span the period band 23 years to 0.4 hour. This paper provides background information and stratigraphy relevant to assessing the significance of the aquifer parameters derived using spectral methods. A regional‐scale three‐dimensional hydrogeological model for the study area is developed, based on the 23‐year record of water level changes that occurred at the various well sites in response to the changing level of Lake Nasser. The model that best fits the data, and is consistent with recent stratigraphic studies of the Nubian formation in the area, consists of three horizontal units of total thickness 400 m. The lowermost unit, overlying the granitic basement, is an aquifer of fluvial sandstones and has a large‐scale (several kilometer) horizontal permeability of 0.32–0.43 μm2 The middle unit is an aquaclude that extends unbroken under the lake and “leaks” at periods longer than several years. Thus the lower aquifer is confined up to periods of several years. The uppermost unit is the water table aquifer, which is composed of 25–30% porosity sandstones with interspersed claystone lenses and has a large‐scale permeability of 1.0–1.5 μm2 Of the wells studied, two are screened in the basal sandstone and four are screened in the water table aquifer. Analysis of tidally and atmospherically forced fluctuations in five of these wells is presented in the companion papers.

Application of time series analysis to circadian rhythms: Effect of beta-adrenergic blockade upon heart rate and transient myocardial ischemia

Circadian variations of transient myocardial ischemia and heart rate have been identified, but the rhythms and their response to β blockade have not been fully characterized. Time-series analysis, a mathematical technique to describe oscillatory activity occurring within a continuous data set was used, to address these issues. Nine men with coronary artery disease underwent 72 hours of ambulatory electrocardiographic monitoring during therapy with placebo or metoprolol. During administration of placebo, ischemic time and heart rate showed a primary peak with a periodicity of approximately 24 hours with a tight coupling between the 2 variables and a secondary peak with a periodicity of 5 to 8 hours. During metoprolol therapy, heart rate and ischemic variation were reduced and the 24-hour periodicity for heart rate only remained. The 24-hour periodicity for ischemia was eliminated, but the data with 5- to 8-hour periodicity became the major component of the signal.

Applied Time Series Analysis for Business and Economic Forecasting

Applied Time Series Analysis for Business and Economic Forecasting

Applied Time Series Analysis for Business and Economic Forecasting

Applied Time Series Analysis for Business and Economic Forecasting

Exploring Emotion in Sub-Structural Aspects of Karelian Lament: Application of Time Series Analysis to Digitized Melody

Interest in the aid of computers within musicology has so far mainly been concentrated on (a) documentation and (b) analysis of music as notated. It still remains to take the important step into that vast field of material, problems, and research that could be labeled music as performed, both as regards the physiological, the acoustical, and the perceptive side. Here it seems possible to predict the arrival of a quite new type of research in the near future. One main problem that remains to be solved is the transformation of registered data from analogous to digital form to get numerical values directly printed out or otherwise stored for computer treatment.

Discussion of the Paper by Bruce and Martin

Discussion of the Paper by Bruce and Martin