On Nonlinear Processing of Air Pollution Data
Three methods — DVS plots, attractor reconstruction, and variance analysis of delay vectors — for detecting nonlinearities in time series are compared on an air pollution dataset. For rigour each method is also used on a surrogate dataset, based on a high-order linear fit to the original data. Finally, a comparison of a standard linear analysis to a neural network model analysis of the air pollution dataset is provided.
KeywordsTime Series Prediction Normalize Little Mean Square Local Linear Model Linear Activation Function Nonlinear Activation Function
Unable to display preview. Download preview PDF.
- J. Theiler, P. S. Linsay, and D. M. Rubin, Time Series Prediction: Forecasting the Future and Understanding the Past, ch. Detecting Nonlinearity in Data with Long Coherenee Times, pp. 429–455. Addison Wesley, 1993.Google Scholar
- D. Beule, H. Herzei, E. Uhlmann, J. Kruger, and F. Becker, “Detecting nonlinearities in time series of machining processes” Proceedings of the American Control Conference, pp. 694–698, 1999.Google Scholar
- M. C. Casdagli and A. S. Weigend, Time Series Prediction: Forecasting the Future and Understanding the Past, ch. Exploring the Continuum Between Deterministie and Stoehastie Modeling, pp. 347–366. Addison Wesley, 1993.Google Scholar
- A. A. M. Khalaf and K. Nakayama, “A hybrid nonlinear predictor: Analysis of learning process and predictability for noisy time series” IEICE Trans. Fundamentals, vol. E82-A, no. 8, pp. 1420–1427, 1999.Google Scholar
- I. R. Kremar, D. P. Mandie, and R. J. Foxall, “On predictability of atmospheric pollution time series” in To appear in the Proceedings of the 5th ICANNGA, 2001.Google Scholar