Abstract
This study proposes a relearning process for a prediction model after detecting structural change points. There are three problems with the detection of structural change points in time-series data: (1) how to generate a prediction model, (2) how to detect a structural change point rapidly, and (3) how the prediction model should relearn after detection. This article targets the third problem and proposes five relearning methods and a process that embeds the relearning process in the sequential probability ratio test. Two experiments, one using 20 generated data sets and the other TOPIX, which consists of 1104 time-series data points between 1991 and 2012, show that using past and future data after detecting the structural change points is helpful.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
*: Yahoo! finance japan. http://finance.yahoo.co.jp/ (2012). Accessed 18 Aug 2014
Blostein, S.: Quickest detection of a time-varying change in distribution. IEEE Trans. Inf. Theory 37(4), 1116–1122 (1991)
Box, G., Jenkins, G.: Time Series Analysis. Forecasting and Control. Prentice Hall, Englewood Cliffs (1976)
Brockwell, P., Davis, R.: Introduction to Time Series and Forecasting. Springer, New York (2003)[AQ2]
Brown, R.L., Durbin, J., Evans, J.M.: Techniques for testing the constancy of regression relationships over time. J. R. Stat. Soc. Ser. B (Methodol.) 37(2), 149–192 (1975)
Bruzzone, L., Fernandez Prieto, D.: A Bayesian approach to automatic change detection. Proceedings of the 1999 IEEE International Geoscience and Remote Sensing Symposium pp. 1816–1818, Hamburg, Germany (1999)
Chow, G.: Tests of equality between sets of coefficients in two linear regressions. Econometrica. 28(3), 591–605 (1960)
Chu, C.S., Hornik, K., Kaun, C.M.: Mosum tests for parameter constancy. Biometrika 82, 603–617 (1995)
Han, C., Willet, P., Abraham, D.: Some methods to evaluate the performance of page’s test as used to detect transient signals. IEEE Trans. Signal Process. 47(8), 2112–2127 (1999)
Horvath, L.: The maximum likelihood method for testing changes in the parameters of normal observations. Ann. Stat. 21, 671–680 (1993)
Hulten, G.: Mining time-changing data streams. Proceeding of the 2001 ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 97–106, San Francisco, USA (2001)
Kawano, H., Hattori, T., Nishimatsu, K.: Structural change point detection method of time series using sequential probability ratio test. IEEJ Trans. Electron. Inf. Syst. 128, 583–592 (2008)
Kivinen, J., Smola, A.J., Williamson, R.C.: Online learning with kernels. IEEE Trans. Signal Process 52, 2165–2176 (2004)
Nyblom, J.: Testing for the constancy of parameters over time. J. Am. Stat. Assoc. 84, 223–230 (1989)
Saad, D.: On-Line Learning in Neural Networks. Cambridge University Press, Cambridge (2009)
Shah, R., Krishnaswamy, S., Gaber, M.: Resource-aware very fast k-means for ubiquitous data stream mining. Proceedings of the 2nd International Workshop on Knowledge Discovery in Data Streams, Porto, Portugal (2005)
Takeda, K., Hattori, T., Izumi, T., Kawano, H.: Extended sprt for structural change detection of time series based on a multiple regression model. Artif. Life Robot. 15, 417–420 (2010)
Wald, A.: Sequential Analysis. Wiley, New York (1947)
Zeileis, A.: A unified approach to structural change tests based on ml scores, f statistics, and ols residuals. Econom. Rev. 24, 445–466 (2005)
Zeileis, A., Leisch, F., Hornik, K., Kleiber, C.: Strucchange. An R package for testing for structural change in linear regression models. http://epub.wu.ac.at/1124/. Accessed 18 Aug 2014
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Saga, R., Kaisaku, N., Tsuji, H. (2015). Relearning Process for SPRT in Structural Change Detection of Time-Series Data. In: Abou-Nasr, M., Lessmann, S., Stahlbock, R., Weiss, G. (eds) Real World Data Mining Applications. Annals of Information Systems, vol 17. Springer, Cham. https://doi.org/10.1007/978-3-319-07812-0_7
Download citation
DOI: https://doi.org/10.1007/978-3-319-07812-0_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-07811-3
Online ISBN: 978-3-319-07812-0
eBook Packages: Business and EconomicsBusiness and Management (R0)