Abstract
This paper proposes an extension of Periodic AutoRegressive (PAR) modelling for time series with evolving features. The large scale of modern datasets, in fact, implies that the time span may subtend several evolving patterns of the underlying series, affecting also seasonality. The proposed model allows several regimes in time and a possibly different PAR process with a trend term in each regime. The means, autocorrelations and residual variances may change both with the regime and the season, resulting in a very large number of parameters. Therefore as a second step we propose a grouping procedure on the PAR parameters, in order to obtain a more parsimonious and concise model. The model selection procedure is a complex combinatorial problem, and it is solved basing on genetic algorithms that optimize an information criterion. The model is tested in both simulation studies and real data analysis from different fields, proving to be effective for a wide range of series with evolving features, and competitive with respect to more specific models.
Similar content being viewed by others
Notes
Downloaded from http://dati-congiuntura.istat.it/.
Downloaded from http://www.metoffice.gov.uk/hadobs/hatcet/data/download.html.
Downloaded from http://www.stats.uwo.ca/faculty/mcleod/epubs/mhsets/MHSETS.zip.
References
Bai, J., Perron, P.: Estimating and testing linear models with multiple structural changes. Econometrica 66(1), 47–78 (1998)
Baragona, R., Battaglia, F., Cucina, D.: Fitting piecewise linear threshold autoregressive models by means of genetic algorithms. Comput. Stat. Data Anal. 47(2), 277–295 (2004)
Baragona, R., Battaglia, F., Poli, I.: Evolutionary Statistical Procedures. Springer, Berlin (2011)
Battaglia, F., Protopapas, M.K.: Multi-regime models for nonlinear nonstationary time series. Comput. Stat. 27(2), 319–341 (2012)
Battaglia, F., Cucina, D., Rizzo, M.: A generalization of periodic autoregressive models for seasonal time series. Tech. Rep. 2, Department of Statistical Sciences, University La Sapienza, Rome, Italy, ISSN 2279-798X (2018)
Bodo, G., Cividini, A., Signorini, L.F.: Forecasting the italian industrial production index in real time. J. Forecast. 10(3), 285–299 (1991)
Box, G.E.P.: Science and statistics. J. Am. Stat. Assoc. 71(356), 791–799 (1976)
Brown, R.L., Durbin, J., Evans, J.M.: Techniques for testing the constancy of regression relationships over time. J. R. Stat. Soc. Ser. B 37(2), 149–192 (1975)
Bruno, G., Lupi, C.: Forecasting industrial production and the early detection of turning points. Empir. Econ. 29(3), 647–671 (2004)
Bulligan, G., Golinelli, R., Parigi, G.: Forecasting monthly industrial production in real-time: from single equations to factor-based models. Empir. Econ. 39(2), 303–336 (2010)
Caporale, G., Cunado, J., Gil-Alana, L.A.: Deterministic versus stochastic seasonal fractional integration and structural breaks. Stat. Comput. 22(2), 349–358 (2012)
Chatterjee, S., Laudato, M., Lynch, L.A.: Genetic algorithms and their statistical applications: an introduction. Comput. Stat. Data Anal. 22(6), 633–651 (1996)
Cucina, D., Rizzo, M., Ursu, E.: Multiple changepoint detection for periodic autoregressive models with an application to river flow analysis (2018). arXiv:1801.01697
Davis, R.A., Lee, T.C.M., Rodriguez-Yam, G.A.: Structural break estimation for nonstationary time series models. J. Am. Stat. Assoc. 101(473), 223–239 (2006)
Davis, R.A., Lee, T.C.M., Rodriguez-Yam, G.A.: Break detection for a class of nonlinear time series models. J. Time Ser. Anal. 29(5), 834–867 (2008)
Falkenauer, E.: Genetic Algorithms and Grouping Problems. Wiley, Chichester (1998)
Franses, P.H.: The effects of seasonally adjusting a periodic autoregressive process. Comput. Stat. Data Anal. 19(6), 683–704 (1995)
Franses, P.H., Paap, R.: Periodic Time Series Models. Oxford University Press, New York (2004)
Girardi, A., Guardabascio, B., Ventura, M.: Factor-augmented bridge models (FABM) and soft indicators to forecast italian industrial production. J. Forecast. 35(6), 542–552 (2016)
Gladyshev, E.: Periodically correlated random sequence. Sov. Math. 2, 385–388 (1961)
Hipel, K.W., McLeod, A.I.: Time Series Modelling of Water Resources and Environmental Systems. Elsevier, Amsterdam (1994)
Holland, J.H.: Adaptation in Natural and Artificial Systems, 2nd edn. MIT Press, Cambridge (1992)
Hubert, L., Arabie, P.: Comparing partitions. J. Classif. 2(1), 193–218 (1985)
Kapetanios, G.: Variable selection in regression models using nonstandard optimisation of information criteria. Comput. Stat. Data Anal. 52(1), 4–15 (2007)
Kleiber, C., Hornik, K., Leisch, F., Zeileis, A.: strucchange: an R package for testing for structural change in linear regression models. J. Stat. Softw. 7(2), 1–38 (2002)
Kontoghiorghes, E.J.: Handbook of Parallel Computing and Statistics. Chapman & Hall/CRC, Boca Raton (2005)
Li, S., Lund, R.: Multiple changepoint detection via genetic algorithm. J. Clim. 25, 674–686 (2012)
Lu, Q., Lund, R.B., Lee, T.C.M.: An MDL approach to the climate segmentation problem. Ann. Appl. Stat. 4(1), 299–319 (2010)
Noakes, D.J., McLeod, A.I., Hipel, K.W.: Forecasting monthly riverflow time series. Int. J. Forecast. 1(2), 179–190 (1985)
Ploberger, W., Krämer, W.: The CUSUM test with OLS residuals. Econometrica 60(2), 271–285 (1992)
Proietti, T., Hillebrand, E.: Seasonal changes in Central England temperatures. J. R. Stat. Soc. Ser A 180(3), 769–791 (2017)
Rand, W.M.: Objective criteria for the evaluation of clustering methods. J. Am. Stat. Assoc. 66(336), 846–850 (1971)
Reeves, C.R.: Modern Heuristic Techniques for Combinatorial Problems. Wiley, New York (1993)
Sin, C.Y., White, H.: Information criteria for selecting possibly misspecified parametric models. J. Econom. 71(1–2), 207–225 (1996)
So, M.K.P., Chung, R.S.W.: Dynamic seasonality in time series. Comput. Stat. Data Anal. 70, 212–226 (2014)
Teräsvirta, T.: Specification, estimation and evaluation of smooth transition autoregressive models. J. Am. Stat. Assoc. 89(425), 208–218 (1994)
Thompstone, R.M., Hipel, K.W., McLeod, A.I.: Grouping of periodic autoregressive models. In: Anderson, O.D., Ord, J.K., Robinson, E.A. (eds.) Time Series Analysis: Theory and Practice 6. Proceedings Conference Toronto 1983, North-Holland, Amsterdam, pp. 35–49 (1985)
Tong, H.: Non Linear Time Series: A Dynamical System Approach. Oxford University Press, Oxford (1990)
Ursu, E., Turkman, K.F.: Periodic autoregressive model identification using genetic algorithms. J. Time Ser. Anal. 33(2), 398–405 (2012)
Wong, H., Ip, W.C., Zhang, R., Xia, J.: Non-parametric time series models for hydrological forecasting. J. Hydrol. 332(3–4), 337–347 (2007)
Zeileis, A., Kleiber, C., Krämer, W., Hornik, K.: Testing and dating of structural changes in practice. Comput. Stat. Data Anal. 44(1–2), 109–123 (2003)
Acknowledgements
We thank the Associate Editor and the Referees for their useful comments and suggestions. We are grateful to Eugen Ursu for involving us in research on periodic autoregressive models. This research was supported by University La Sapienza, Rome under grant C26A15NK4H.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix: Asymptotic equivalence of a test of hypothesis and an information criterion
Appendix: Asymptotic equivalence of a test of hypothesis and an information criterion
Let us consider two nested models \(M_0\) and \(M_1\), where \(M_0\) has p parameters and \(M_1\) has q additional parameters that under the null hypothesis \(H_0\) are assumed equal to zero. A standard F test under gaussianity for \(H_0\) is obtained by the test statistic:
where \(SS_0\) and \(SS_1\) are the residual sum of squares of the model \(M_0\) and \(M_1\) fitted to N observations.
Under the null hypothesis the statistic (4) follows a F distribution with q and \(N-p-q\) degrees of freedom. Let \({\overline{F}} _\alpha \) denote the \((1-\alpha )\)-quantile of that distribution, then hypothesis \(H_0\) is rejected at level \(1-\alpha \) whenever \(F>{\overline{F}}_\alpha \), or
For N large the right hand side may be approximated by \({\overline{F}}_\alpha \, q/(N-p-q)\) and rejection is equivalent to
that may be also written \(\mathrm{IC}_0^* - \mathrm{IC}_1^* >0\) where \(\mathrm{IC}_0^*= \log (SS_0) + \pi ^* \, p\) , \(\mathrm{IC}_1^*= \log (SS_1) + \pi ^* \, (p+q)\) and
It follows that choosing a penalizing constant equal to \(\pi ^*\) in (5), model \(M_1\) will be preferred to model \(M_0\) by the information criterion whenever the test rejects the null hypothesis \(H_0\).
Rights and permissions
About this article
Cite this article
Battaglia, F., Cucina, D. & Rizzo, M. Parsimonious periodic autoregressive models for time series with evolving trend and seasonality. Stat Comput 30, 77–91 (2020). https://doi.org/10.1007/s11222-019-09866-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-019-09866-0