Plusieurs données de marché, telles que les prix spot de l'électricité, présentent de la longue mémoire, au sens de la décroissance hyperbolique des autocorrélations combinée avec un phénomène d'hétéroskédasticité. Pour modéliser de tels comportements, nous considérons dans cette Note les processus GIGARCH à k facteurs et nous proposons deux méthodes d'estimation des paramètres de ce modèle. Enfin, nous développons les propriétés asymptotiques de ces estimateurs.
Some crucial time series of market data, such as electricity spot prices, exhibit long-memory, in the sense of slowly-decaying correlations combined with heteroskedasticity. To be able to modelize such a behaviour, we consider in this Note the k-factor GIGARCH process and we propose two methods to address the related parameter estimation problem. For each method, we develop the asymptotic theory for the estimation.
Accepté le :
Publié le :
@article{CRMATH_2004__339_6_435_0, author = {Diongue, Abdou K\^a and Gu\'egan, Dominique}, title = {Estimating parameters of a \protect\emph{k}-factor {GIGARCH} process}, journal = {Comptes Rendus. Math\'ematique}, pages = {435--440}, publisher = {Elsevier}, volume = {339}, number = {6}, year = {2004}, doi = {10.1016/j.crma.2004.07.014}, language = {en}, url = {http://www.numdam.org/articles/10.1016/j.crma.2004.07.014/} }
TY - JOUR AU - Diongue, Abdou Kâ AU - Guégan, Dominique TI - Estimating parameters of a k-factor GIGARCH process JO - Comptes Rendus. Mathématique PY - 2004 SP - 435 EP - 440 VL - 339 IS - 6 PB - Elsevier UR - http://www.numdam.org/articles/10.1016/j.crma.2004.07.014/ DO - 10.1016/j.crma.2004.07.014 LA - en ID - CRMATH_2004__339_6_435_0 ER -
%0 Journal Article %A Diongue, Abdou Kâ %A Guégan, Dominique %T Estimating parameters of a k-factor GIGARCH process %J Comptes Rendus. Mathématique %D 2004 %P 435-440 %V 339 %N 6 %I Elsevier %U http://www.numdam.org/articles/10.1016/j.crma.2004.07.014/ %R 10.1016/j.crma.2004.07.014 %G en %F CRMATH_2004__339_6_435_0
Diongue, Abdou Kâ; Guégan, Dominique. Estimating parameters of a k-factor GIGARCH process. Comptes Rendus. Mathématique, Tome 339 (2004) no. 6, pp. 435-440. doi : 10.1016/j.crma.2004.07.014. http://www.numdam.org/articles/10.1016/j.crma.2004.07.014/
[1] Asymptotic properties of maximum likelihood estimators for stochastic processes, Sankhya Ser. A, Volume 38 (1976), pp. 259-270
[2] Estimating a generalized long-memory process, J. Econometrics, Volume 73 (1996), pp. 237-259
[3] A generalized fractionally integrated ARMA process, J. Time Series Anal., Volume 17 (1994), pp. 111-140
[4] A.K. Diongue, D. Guégan, B. Vignal, Processus GIGARCH: Estimation et applications aux prix spot de l'électricité, Preprint MORA, 14, 2003
[5] Comparison of parameter estimation methods in cyclical long-memory time series (Dunis, C.; Timmermann, J., eds.), Development in Forecasts Combination and Portfolio Choice, Wiley, 2001
[6] Whittle estimation of ARCH models, Econometrics Theory, Volume 17 (2001), pp. 608-631
[7] A new model: The k-factor GIGARCH process, J. Signal Process., Volume 4 (2000), pp. 265-271
[8] A prospective study of the k-factor Gegenbauer process with heteroscedastic errors and an application to inflation rates, Finance India, Volume 17 (2003), pp. 1-20
[9] A Limit theory for long-range dependence and statistical inference on related models, Ann. Stat., Volume 25 (1997), pp. 105-137
[10] Almost Sure Convergence, Academic Press, New York, 1974
Cité par Sources :