Parameter estimation and their change-point detectionfor generalized autoregressive process with conditional heteroscedasticity
The problem of change-point detection of the parameters of GARCH(p,q) process is considered.The utoregressive parameters of the process before and after the change point are supposedto be unknown. A sequential procedure for estimating the parameters based on the weighted leastsquares method is developed. The choice of the weights and the stopping rule allows one to constructan estimator with a preassigned mean square error depending on parameter H of the procedure.The procedure of change-point detection is based on comparison of parameter estimators ondifferent observation intervals. The upper bounds for probabilities of the false alarm and the delayare found.
Download file
Counter downloads: 346
Keywords
mean square error, guaranteed estimation, least squares method, GARCH(p,q), change-point, гарантированное оценивание, среднеквадратическое отклонение, метод наименьших квадратов, момент разладки, GARCH(p,q)Authors
| Name | Organization | |
| Burkatovskaya Yulia B. | National Research Tomsk Polytechnic University | tracey@tpu.ru |
| Vorobeychikov Sergey E. | National Research Tomsk State University | sev@mail.tsu.ru |
| Sergeeva Ekaterina E. | National Research Tomsk Polytechnic University | sergeeva_e_e@mail.ru |
References
Дмитренко А.А., Конев В.В. О последовательной классификации процессов авторегрессии с неизвестной дисперсией помех // Проблемы передачи информации. 1995. Т. 31. Вып. 4. С. 51−62.
Meder N., Vorobejchikov S. On guaranteed estimation of parameters of random processes by the weighted least square method // Proc. 15th Triennial World Congress of the International Federation of Automatic Control. Barcelona, 2002. No. 1200.
Baillie R.T., Chung H. Estimation of GARCH models from the autocorrelation of the squares of a process // J. Time Ser. Anal. 2001. V. 22. No. 6. P. 631−650.
Storti G. Minimum distance of GARCH(1,1) models // Computattional Statistics and Data Analysis. 2006. V. 51. P. 1803−1821.
Буркатовская Ю. Б., Воробейчиков С.Э. Гарантированное обнаружение момента разладки GARCH-процесса // Автоматика и телемеханика. 2006. № 12. С. 56−70.
Peng I., Yao Q. Least absolute deviations estimation for ARCH and GARCH models // Biometrica. 2003. V. 90. N. 4. P. 967−997.
Muler N., Yohai V.J. Robust estimates for GARCH models // J. Statistical Planning and Inference. 2008. V. 138. N. 10. P. 2918−2940.
Boudt K., Croux C. Robust M-estimation of multivariate GARCH models // Computational Statistics and Data Analysis. 2010. V. 54. P. 2459−2469.
Hamadeh T., Zakoian J.-M. Asymptotic properties of LS and QML estimators for a class of nonlinear GARCH processe // J. Statist. Plann.Inference. 2011. V. 141. P. 488−507.
Bai J. Least squares estimation of a shift in linear process // J. Time Series Analysis. 1994. V. 15. N.5. P. 453−472.
Francq C., Zacoian J.M. Maximum likelihood estimation of pure GARCH and ARMA - GARCH processes // Bernoulli. 2004. V. 10. P. 605−637.
Pan J., Wang H., Tong H. Estimation and power-transformed and threshold GARCH models // J. Econometrics. 2008. V. 142. P. 352−378.
Berkes I., Horvath L., Kokoszka P.S. GARCH processes: Structure and estimation // Bernoulli. 2003. V. 9. P. 201−227.
Berkes I., Horvath L. The efficiency of the estimators of the parameters in GARCH processes // Annals of Statistics. 2004. V. 32. P. 633−655.
Gombey E., Serban D. Monitoring parameter change in AR(p) time series models // Statistics Centre Technical Reports 05.04, The University of Alberta, Edmonton, Canada, 2005.
Gombey E. Change detection in autoregressive time series // J. Multivariate Analysis. 2008. V. 99. P. 451−464.
Hillebrand E. Negleting parameter changes in GARCH models// J. Econometrics. 2005. V. 129. P. 121−138.
Davies R.A., Huang D., Yao Y.-C. Testing for change in the parameter value and order of autoregressive model // Ann. Statist. 1995. V. 23. P. 282−304.
Bollerslev T. Generalized autoregressive conditional heteroskedasticity // J. Econometrics. 1986. V. 86. P. 307−327.
