ERU - 2005

Permanent URI for this collectionhttp://192.248.9.226/handle/123/14682

Browse

Browsing ERU - 2005 by Author "Cooray, TMJA"

Filter results by typing the first few letters
Now showing 1 - 3 of 3
  • Thumbnail Image
    item: Conference-Full-text
    An approach of state space modelling for indirect tire pressure monitoring
    (2005) Cooray, TMJA
    In this paper, dynamical vehicle models and Time dependent State Space model have successfully been used to detect tire pressure losses(non observabilty). When the pressure is decreased in a tire, the corresponding wheel radius is reduced and the angular velocity increases. The basic idea behind this paper, is to indirectly estimate the wheel radius via secondary existing sensors. Two dynamical vehicle models, in which the relative wheel radius is included as model parameters, are formed. With these models any combination of pressure loss in the tires can be detected
  • Thumbnail Image
    item: Conference-Full-text
    Estimation and forecasting of time dependent parameter models using reliable software
    (2005) Cooray, TMJA
    Until recently, the dominant paradigm in the analysis and forecasting of nonstationary time series has been the approach proposed originally by Box and Jenkins in 1970, which involves the en bloc processing of time series data that have been reduced to stationarity by pre-processing, using techniques such as differencing and use of transformation. A more flexible and widely applicable alternative, which is now favored in many different scientific disciplines, is to analyse the time series directly in their non stationary form using recursive estimation and fixed interval smoothing. Here, the estimates of model parameters or state variables are updated sequentially, so allowing for the estimation of the time variable or state dependent parameters that can be used to characterise models of nonstationary systems. This paper provides an introduction to the latest techniques, developed by the author using MATLAB tool box, in optimal recursive estimation and concentrates on the simplest class of models for nonstationary systems; namely time variable parameter, or Vector Auto Regressive Moving Averages with eXogenous variables (VARMAX), Generalised AutoRegressive Conditional Heteroscedastic (GARCH) errors, as well as the closely related time variable parameter version of the State Space time dependant (SDP) models. In all cases, the utility of these methods is demonstrated through examples
  • Thumbnail Image
    item: Conference-Full-text
    A time-dependant parameter approach of demand analysis of tourism in Sri Lanka
    (2005) Cooray, TMJA
    Traditional tourism demand analysis uses ordinary least squares or maximum likelihood methods to estimate demand models like Box-Jenkins and State-Space, assuming that the parameters of the models remain constant over the sample period. This assumption is too restrictive, as it does not allow for behavioral changes of arrival of tourists over time. This study proposes a new methodology the Generalized autoregressive conditional heterosedasticity model (GARCH) approach to tourism demand modeling. This method relaxes the assumption of parameter constancy, and the behavioral change of tourists over time is traced using a statistical estimator known as a Kalman filter. GARCH models permit time-varying conditional covariances as well as variances, and the former quantity can be of substantial practical use for both modeling and forecasting. The appropriateness of the GARCH approach to tourism demand modeling is tested based on a data set of the tourist demand for Sri Lanka and estimated Mean Percentage Errors(MAPE) are explained 9.7% 6% and 2% respectively