Continuous Time State Space Modelling with an Application to High-Frequency Road Traffic Data
We review Kalman filter and related smoothing methods for the continuous time state space model. The attractive property of continuous time state space models is that time gaps between consecutive observations in a time series are allowed to vary throughout the process. We discuss some essential details of the continuous time state space methodology and review the similarities and the differences between the continuous time and discrete time approaches. An application in the modelling of road traffic data is presented in order to illustrate the relevance of continuous time state space modelling in practice.
We thank Rijkswaterstaat, The Netherlands (WVL), for providing us with the data set.
- Bergstrom, A. R. (1984). Gaussian estimation of structural parameters in higher order continuous time dynamic models. In Z. Griliches & M. Intriligator (Eds.), The handbook of econometrics (Vol. 2, pp. 1145–1212). Amsterdam: North-Holland. https://doi.org/10.1016/s1573-4412(84)02012-2.Google Scholar
- Doornik, J. A. (2013). Object-oriented matrix programming using Ox 7.00. London: Timberlake Consultants Press.Google Scholar
- Durbin, J., & Koopman, S. J. (2012). Time series analysis by state space methods (2nd ed.). Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199641178.001.0001.CrossRefGoogle Scholar
- Harvey, A. C. (1989). Forecasting, structural time series models and the Kalman filter. Cambridge: Cambridge University Press.Google Scholar
- Koopman, S. J., Harvey, A. C., Doornik, J. A., & Shephard, N. (2007). STAMP 8.0: Structural time series analyser, modeller and predictor. London: Timberlake Consultants.Google Scholar
- Wahba, G. (1990). Spline models for observational data. Philadelphia: SIAM. https://doi.org/10.1137/1.9781611970128.