Kybernetika 54 no. 6, 1122-1137, 2018

Change point detection in vector autoregression

Zuzana PráškováDOI: 10.14736/kyb-2018-6-1122

Abstract:

In the paper a sequential monitoring scheme is proposed to detect instability of parameters in a multivariate autoregressive process. The proposed monitoring procedure is based on the quasi-likelihood scores and the quasi-maximum likelihood estimators of the respective parameters computed from a training sample, and it is designed so that the sequential test has a small probability of a false alarm and asymptotic power one as the size of the training sample is sufficiently large. The asymptotic distribution of the detector statistic is established under both the null hypothesis of no change as well as under the alternative that a change occurs.

Keywords:

change point, vector autoregression, quasi-maximum likelihood

Classification:

62M10, 62E20

References:

  1. J. Bai and P. Perron: Estimating and testing linear models with multiple structural changes. Econometrica 66 (1998), 47-78.   DOI:10.2307/2998540
  2. I. Berkes, E. Gombay, L. Horváth and P. Kokoszka: Sequential change-point detection in GARCH(p,q) models. Econometr. Theory 20 (2004), 1140-1167.   DOI:10.1017/s0266466604206041
  3. Y. Boubacar Mainassara and C. Francq: Estimating structural VARMA models with uncorrelated but non-independent error terms. J. Multivariate Anal. 102 (2011), 496-505.   DOI:10.1016/j.jmva.2010.10.009
  4. F. Carsoule and P. H. Franses: A note on monitoring time-varying parameters in an autoregression.    DOI:10.1007/s001840200198
  5. C.-S. J. Chu, M. Stinchcombe and H. White: Monitoring structural change.    DOI:10.2307/2171955
  6. J. Davidson: Stochastic Limit Theory. Oxford University Press, Oxford 1994.   DOI:10.1093/0198774036.001.0001
  7. M. Dvořák: Stability in Autoregressive Time Series Models. Ph.D. Dissertation, Charles University, Prague 2015.   CrossRef
  8. E. Gombay and D. Serban: Monitoring parameter change in AR(p) time series models.    DOI:10.1016/j.jmva.2008.08.005
  9. J. D. Hamilton: Time Series Analysis. Princeton University Press, Princeton 1994.   DOI:10.1017/s0266466600009440
  10. L. Horváth, M. Hušková, P. Kokoszka and J. Steinebach: Monitoring changes in linear models.    DOI:10.1016/j.jspi.2003.07.014
  11. Z. Hlávka, M. Hušková, C. Kirch and S. G. Meintanis: Monitoring changes in the error distribution of autoregressive models based on Fourier methods. Test 21 (2012), 605-634.   DOI:10.1007/s11749-011-0265-z
  12. M. Hušková and A. Koubková: Sequential procedures for detection of changes in autoregressive sequences.    CrossRef
  13. J. Kuelbs and W. Philipp: Almost sure invariance principles for partial sums of mixing $B-$valued random variables.    DOI:10.1214/aop/1176994565
  14. F. Li, Z. Tian and P. Qi: Structural change monitoring for random coefficient autoregressive time series. Comm. Statist. Simulation Comput. 44 (2015), 996-1009.   DOI:10.1080/03610918.2013.800205
  15. S. Lee, Y. Lee and O. Na: Monitoring distributional changes in autoregressive models    DOI:10.1080/03610920902947261
  16. H. Lütkepohl: New Introduction to Multiple Time Series Analysis. Springer-Verlag, Berlin 2005.   DOI:10.1007/978-3-540-27752-1
  17. O. Na, J. Lee and S. Lee: Monitoring parameter changes for random coefficient autoregressive models.    DOI:10.1016/j.jkss.2010.03.006
  18. Z. Prášková: Monitoring changes in RCA models. In: Stochastic Models, Statistics and Their Applications, Springer Proceedings in Mathematics and Statistics 122 (A. Steland, E. Rafajlowicz and K. Szajowski, eds.), Springer, Berlin 2015, pp. 129-137.   DOI:10.1007/978-3-319-13881-7\_15
  19. D. Straumann: Estimation in Conditionally Heteroscedastic Time Series Models. Lecture Notes in Statistics 181, Springer, Berlin 2005.   DOI:10.1007/b138400