[Estimation sous contraintes d'une matrice de covariances dans la modélisation d'accidents de la route par complements de Schur]
On considère (Y11,Y21),…,(Y1s,Y2s) une collection de s couples indépendants de vecteurs aléatoires de dimension 2r avec (r>1). On suppose que chaque couple est distribué selon une loi multinomiale dépendant à la fois d'un vecteur paramètre inconnu et d'un ensemble de données supplémentaire. On étudie l'inversion formelle de l'information de Fisher issue de ces s couples par le biais de la technique des compléments de Schur.
We consider a collection (Y11,Y21),…,(Y1s,Y2s) of s independent couples of 2r-dimensional random vector with r>1. We assume that each couple has a multinomial distribution linked to an unknown vector parameter and an extra set data. We deal with a formal inversion of a Fisher's information matrix connected to those couples using Schur complements approach.
Accepté le :
Publié le :
@article{CRMATH_2003__337_3_219_0,
author = {N'Guessan, Assi},
title = {Constrained covariance matrix estimation in road accident modelling with {Schur} complements},
journal = {Comptes Rendus. Math\'ematique},
pages = {219--222},
publisher = {Elsevier},
volume = {337},
number = {3},
year = {2003},
doi = {10.1016/S1631-073X(03)00309-1},
language = {en},
url = {http://www.numdam.org/articles/10.1016/S1631-073X(03)00309-1/}
}
TY - JOUR AU - N'Guessan, Assi TI - Constrained covariance matrix estimation in road accident modelling with Schur complements JO - Comptes Rendus. Mathématique PY - 2003 SP - 219 EP - 222 VL - 337 IS - 3 PB - Elsevier UR - http://www.numdam.org/articles/10.1016/S1631-073X(03)00309-1/ DO - 10.1016/S1631-073X(03)00309-1 LA - en ID - CRMATH_2003__337_3_219_0 ER -
%0 Journal Article %A N'Guessan, Assi %T Constrained covariance matrix estimation in road accident modelling with Schur complements %J Comptes Rendus. Mathématique %D 2003 %P 219-222 %V 337 %N 3 %I Elsevier %U http://www.numdam.org/articles/10.1016/S1631-073X(03)00309-1/ %R 10.1016/S1631-073X(03)00309-1 %G en %F CRMATH_2003__337_3_219_0
N'Guessan, Assi. Constrained covariance matrix estimation in road accident modelling with Schur complements. Comptes Rendus. Mathématique, Tome 337 (2003) no. 3, pp. 219-222. doi : 10.1016/S1631-073X(03)00309-1. http://www.numdam.org/articles/10.1016/S1631-073X(03)00309-1/
[1] Maximum likelihood estimation of parameters subject to restraints, Ann. Math. Statist., Volume 29 (1958), pp. 813-829
[2] Maximum likelihood estimation for dependent observations, J. Roy. Statist. Soc., Volume 38 (1975), pp. 45-53
[3] On the constrained maximum likelihood estimation with non i.i.d. observations, Ann. Inst. Statist. Math. Part A, Volume 36 (1984), pp. 239-249
[4] Maximum likelihood approaches to variance component estimation and to related problems, J. Amer. Statist. Assoc., Volume 72 (1977) no. 358, pp. 320-340
[5] Constrained estimation in covariance structure analysis, Biometrika, Volume 66 (1979) no. 3, pp. 539-545
[6] Estimation multidimensionnelle des contrôles et de l'effet moyen d'une mesure de sécurité routière, Rev. Statist. Appl., Volume XLIX (2001) no. 2, pp. 83-100
[7] On a use of Schur complements for an inverse of a constrained accident data information matrix, Pub. IRMA Lille, Volume 60 (2003) no. VI
[8] Schur complements and statistics, Linear Algebra Appl., Volume 36 (1981), pp. 187-295
[9] Estimation in restricted parameter spaces – some history and some recent developments, CWI Quarterly, Volume 9 (1 and 2) (1996), pp. 69-76
Cité par Sources :
