|
Conference publicationsAbstractsXV conferenceInformation matrix for DCC-MGARCH(1,1)Tomsk Polytechnic University, Faculty of Natural Science and Mathematics, Department of Higher Mathematics and Mathematical Physics, 634012, Russia, Tomsk, Lenin av., 30, Tel. (3822) 41-98-13, fax: (3822) 56-39-08, Е-mail: belsner@bk.ru, olegkol@tpu.ru 1 pp.The analytic form of Fisher Information Matrix (IM) written for DCC-MGARCH(1,1) was suggested. After that, it was applied for simplifying the general algorithm: the statistical hypothesis about constant correlation matrix usage was put forward and statistical verification was made. IM was employed for Russian share market: to do investigations the five equilibrium portfolios was compounded from four different shares in each case. Computations made showed that there are three types T1–T3 of trading days on the market and day type changing from T1 to T2 and vice versa are happening over the time moments T3. Moreover, the clusterization effect of multivariate volatility that investigated by scientists from all around the world in the univariate case, was discovered and described. References. 1R.F Engle, Dynamic conditional correlation – a simple class of multivariate GARCH models//Journal of Business and Economic Statistic, V. 20, 2002, p. 339 – 350. 2.T. Bollerslev, Modelling the Coherence in Short-Run Nominal Exchange Rates: A Multivariate Generalized ARCH model//The Review of Economics and Statistics, V.72, №3, 1990, p. 498–505.
|
