Mesure de risque : détection du régime de crise et calcul de la Value-at-Risk

Nicolas Champagnat (1, 2) , Madalina Deaconu (1, 2) , Antoine Lejay (1, 2) , Khaled Salhi (1, 2)

1 TOSCA - TO Simulate and CAlibrate stochastic models
2 IECL - Institut Élie Cartan de Lorraine

This work aims to study the behavior of financial assets' returns and to measure market risk. Using the hidden Markov model, we classify the data according to the criteria of crisis by detecting the crisis en non-crisis regimes. By adopting the power laws to model the tail of the distribution and by taking into account this classification, we measure the Value-at-Risk (VaR) and we propose a new indicator for portfolio management.

Mots clés: hidden Markov models; Hill estimator; financial mathematics; value-at-risk; power law

Domaines: Probabilités [math.PR]

Nicolas Champagnat, Madalina Deaconu, Antoine Lejay, Khaled Salhi. Mesure de risque : détection du régime de crise et calcul de la Value-at-Risk. [Contrat] non précisé. 2013, pp.67. ⟨hal-00942009⟩