Tests of the Constancy of Conditional Correlations of Unknown Functional Form in Multivariate GARCH Models
Fiche du document
2016
- handle: 10670/1.t72xr7
- hal: hal-04218472
- doi: 10.15609/annaeconstat2009.123-124.0077
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.15609/annaeconstat2009.123-124.0077
info:eu-repo/semantics/OpenAccess
Mots-clés
Multivariate GARCH Neural Network Taylor Expansion C - Mathematical and Quantitative Methods/C.C2 - Single Equation Models • Single Variables/C.C2.C22 - Time-Series Models • Dynamic Quantile Regressions • Dynamic Treatment Effect Models • Diffusion Processes C - Mathematical and Quantitative Methods/C.C4 - Econometric and Statistical Methods: Special Topics/C.C4.C45 - Neural Networks and Related Topics C - Mathematical and Quantitative Methods/C.C5 - Econometric Modeling/C.C5.C58 - Financial EconometricsSujets proches
Competitive examinations TestsCiter ce document
Anne Peguin-Feissolle et al., « Tests of the Constancy of Conditional Correlations of Unknown Functional Form in Multivariate GARCH Models », HAL-SHS : économie et finance, ID : 10.15609/annaeconstat2009.123-124.0077
Métriques
Partage / Export
Résumé
We introduce two tests for the constancy of conditional correlations of unknown functional form in multivariate GARCH models. The first test is based on artificial neural networks and the second on a Taylor expansion of each unknown conditional correlation. They can be seen as general misspecification tests for a large set of multivariate GARCH-type models. We investigate their size and their power through Monte Carlo experiments. Moreover, we study the robustness of these tests to nonnormality by simulating some models, such as the GARCH − t and Beta − t − EGARCH . We give some illustrative empirical examples based on financial data.