Abstract
This paper addresses issue of modeling, analysis and forecasting of time series drifted by autoregressive noise and finding its optimal solution by extending a conventional linear growth model with an autoregressive component. This additional component is designed to take care of high frequencies of autoregressive noise drift without influencing the low frequencies of the linear trend and compromising on parsimonious nature of the model. The parameters of this model are then optimally estimated through the self updating recursive equations using Bayesian priors. For identification of autoregressive order of noise and estimation of its coefficients ATS procedure of Akram (2001) is employed. Further, for unknown variance of observations an on-line variance learning and estimation procedure is discussed. To demonstrate practical aspects of the model some examples are given and for generation of short, medium and long term forecasts in one go an appropriate forecast function is given.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Akaike H (1973) Information theory and an extension of the maximum likelihood principle. In: Proceedings of the 2nd International Symposium on Inference Theory. BN Petran, F. Csáki Eds. Akadémiai Kiadi, Budapest, Hungary, pp 267–281
Akram M (1988) Recursive transformation matrices for linear dynamic system models. J Computational Stat & Data Analysis 6:119–127
Akram M (1992) Construction of state space models for time series exhibiting exponential growth. In: Computational Statistics, vol 1, Physica Verlag, Heidelberg, Germany, pp 303–308
Akram M (1994) Computational aspects of state space models for time series forecasting. Proceedings of 11th Symposium on Computational Statistics (COMPSTAT-1994), Vienna, Austria, pp 116–117
Akram M (2001) A test statistic for identification of noise processes. Pakistan Journal of Statistics 17(2):103–115
Akram M, Irfan A (2007) Identification of optimum statistical models for time series analysis and forecasting using akaike information criterion and akram test statistic: A comparative study. Proc.of World Congress of Engineers, London, vol 2, pp 956–960
Bohlin T (1978) Maximum-power validation of models without higher-order fitting. Automatica 14:137–146
Harrison P, Akram M (1983) Generalized exponentially weighted regression and parsimonious dynamic linear modelling. Time Series Analysis: Theory and Practice 3:102–139
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer London
About this paper
Cite this paper
Chaudhry, A. (2009). Parsimonious Modeling and Forecasting of Time Series drifted by Autoregressive Noise. In: Reiner, G. (eds) Rapid Modelling for Increasing Competitiveness. Springer, London. https://doi.org/10.1007/978-1-84882-748-6_4
Download citation
DOI: https://doi.org/10.1007/978-1-84882-748-6_4
Publisher Name: Springer, London
Print ISBN: 978-1-84882-747-9
Online ISBN: 978-1-84882-748-6
eBook Packages: EngineeringEngineering (R0)