Home  About us  Contact
 

Series Techniques (series + techniques)

Distribution by Scientific Domains
Mathematics and Statistics50%

Kinds of Series Techniques

  • time series techniques
    		•

    Selected Abstracts

    AN ASSESSMENT OF RECENT TRENDS IN GIRLS' VIOLENCE USING DIVERSE LONGITUDINAL SOURCES: IS THE GENDER GAP CLOSING?

    CRIMINOLOGY, Issue 2 2005
    DARRELL STEFFENSMEIER
    Applying Dickey-Fuller time series techniques in tandem with intuitive plot-displays, we examine recent trends in girls' violence and the gender gap as reported in four major sources of longitudinal data on youth violence. These sources are arrest statistics of the Uniform Crime Reports, victimization data of the National Crime Victimization Survey (where the victim identifies sex of offender) and self-reported violent behavior of Monitoring the Future and National Youth Risk Behavior Survey. We find that the rise in girls' violence over the past one to two decades as counted in police arrest data from the Uniform Crime Reports is not borne out in unofficial longitudinal sources. Several net-widening policy shifts have apparently escalated girls' arrest-proneness: first, stretching definitions of violence to include more minor incidents that girls in relative terms are more likely to commit; second, increased policing of violence between intimates and in private settings (for example, home, school) where girls' violence is more widespread; and, third, less tolerant family and societal attitudes toward juvenile females. These developments reflect both a growing intolerance of violence in the law and among the citizenry and an expanded application of preventive punishment and risk management strategies that emphasize early identification and enhanced formal control of problem individuals or groups, particularly problem youth. [source]

    Marketing Category Forecasting: An Alternative of BVAR-Artificial Neural Networks¶

    DECISION SCIENCES, Issue 4 2000
    James J. Jiang
    ABSTRACT Analyzing scanner data in brand management activities presents unique difficulties due to the vast quantity of the data. Time series methods that are able to handle the volume effectively often are inappropriate due to the violation of many statistical assumptions in the data characteristics. We examine scanner data sets for three brand categories and examine properties associated with many time series forecasting methods. Many violations are found with respect to linearity, normality, autocorrelation, and heteroscedasticity. With this in mind we compare the forecasting ability of neural networks that require no assumptions to two of the more robust time series techniques. Neural networks provide similar forecasts to Bayesian vector autoregression (BVAR), and both outperform generalized autoregressive conditional herteroscedasticty (GARCH) models. [source]

    Estimating common trends in multivariate time series using dynamic factor analysis

    ENVIRONMETRICS, Issue 7 2003
    A. F. Zuur
    Abstract This article discusses dynamic factor analysis, a technique for estimating common trends in multivariate time series. Unlike more common time series techniques such as spectral analysis and ARIMA models, dynamic factor analysis can analyse short, non-stationary time series containing missing values. Typically, the parameters in dynamic factor analysis are estimated by direct optimization, which means that only small data sets can be analysed if computing time is not to become prohibitively long and the chances of obtaining sub-optimal estimates are to be avoided. This article shows how the parameters of dynamic factor analysis can be estimated using the EM algorithm, allowing larger data sets to be analysed. The technique is illustrated on a marine environmental data set. Copyright © 2003 John Wiley & Sons, Ltd. [source]

    A step-wise incremented Fourier series model for chemical reactivity prediction

    JOURNAL OF CHEMOMETRICS, Issue 6-7 2006
    Saloua Saidane
    Abstract In this paper, chemical reactivity is modeled as a time series of events defined by a reactant's concentration decay measured at consecutive discrete time periods. Since traditional time series techniques such as ARIMA and current Artificial Neural Networks require large data sets that are typically not available for chemical reactions, we developed a Step Wise Incremented Fourier Series (SWIFS) algorithm to model and predict nonlinear short time series. The application of SWIFS to experimental data from first- and second-order reactions produced a significant improvement in prediction accuracy over traditional integrated rate laws. In forward-time prediction, SWIFS has achieved significantly higher prediction accuracy with first- and second-order chemical reactions data. SWIFS also proved more robust in terms of error propagation caused by the effect of the size of the estimation set. The proposed SWIFS model also outperformed rate law models in backwards-time prediction. The ability of SWIFS to provide high accuracy in predicting chemical reactions may have beneficial implications on the efficiency of industrial production of chemicals as well as on the effective control of hazardous materials degradation. Copyright © 2007 John Wiley & Sons, Ltd. [source]

    Nonparametric smoothing using state space techniques

    THE CANADIAN JOURNAL OF STATISTICS, Issue 1 2001
    Patrick E. Brown
    Abstract The authors examine the equivalence between penalized least squares and state space smoothing using random vectors with infinite variance. They show that despite infinite variance, many time series techniques for estimation, significance testing, and diagnostics can be used. The Kalman filter can be used to fit penalized least squares models, computing the smoothed quantities and related values. Infinite variance is equivalent to differencing to stationarity, and to adding explanatory variables. The authors examine constructs called "smoothations" which they show to be fundamental in smoothing. Applications illustrate concepts and methods. Les auteurs examinent l'équivalence entre les moindres carrés pénalisés et le lissage de l'espace d'états au moyen de vecteurs aléatoires à variance infinie. Ils montrent que malgré le problème de variance infinie, plusieurs techniques de diagnostic, d'estimation et de test de signification propres aux chroniques restent valables. Le filtre de Kalman permet d'évaluer les modèles des moindres carrés pénalisés en fournissant entre autres des valeurs lissées. La variance infinie est équivalente à la différenciation à des fins de stationnarité et à l'ajout de variables explicatives. Les auteurs étudient en outre des quantités appelées "lissations," dont ils montrent l'importance pour le lissage. Des applications illustrent les méthodes et procédures décrites. [source]

    A multivariate time series approach to projected life tables

    APPLIED STOCHASTIC MODELS IN BUSINESS AND INDUSTRY, Issue 6 2009
    Dorina Lazar
    Abstract The method of mortality forecasting proposed by Lee and Carter describes a time series of age-specific log-death rates as a sum of an independent of time age-specific component and a bilinear term in which one of the component is a time-varying factor reflecting general change in mortality and the second one is an age-specific parameter. Such a rigid model structure implies that on average the mortality improvements for different age groups should be proportional, regardless of the calendar period: a single time factor drives the future death rates. This paper investigates the use of multivariate time series techniques for forecasting age-specific death rates. This approach allows for relative speed of decline in the log death rates specific to the different ages. The dynamic factor analysis and the Johansen cointegration methodology are successfully applied to project mortality. The inclusion of several time factors allows the model to capture the imperfect correlations in death rates from 1 year to the next. The benchmark Lee,Carter model appears as a special case of these approaches. An empirical study is conducted with the help of the Johansen cointegration methodology. A vector-error correction model is fitted to Belgian general population death rates. A comparison is performed with the forecast of life expectancies obtained from the classical Lee,Carter model. Copyright © 2009 John Wiley & Sons, Ltd. [source]