A new particle filtering approach to estimate stochastic volatility models with Markov-switching - HAL Accéder directement au contenu
Article dans une revue Econometrics and Statistics Année : 2018

A new particle filtering approach to estimate stochastic volatility models with Markov-switching

Résumé

A simple method is proposed to estimate stochastic volatility models with Markov-switching. It relies on a nested structure of filters (a Hamilton filter and several particle filters) to approximate unobserved regimes and state variables, respectively. Smooth resampling is used to keep the computational complexity constant over time and to implement a standard likelihood-based inference on parameters. A bootstrap and an adapted version of the filter are described and their performance are assessed using simulation experiments. The volatility of US and French markets is characterized over the last decade using a three-regime stochastic volatility model extended to include a leverage effect.
Loading...
Fichier non déposé

Dates et versions

hal-02296093, version 1 (24-09-2019)

Identifiants

Citer

Frédéric Karamé. A new particle filtering approach to estimate stochastic volatility models with Markov-switching. Econometrics and Statistics , 2018, 8, pp.204-230. ⟨10.1016/j.ecosta.2018.05.004⟩. ⟨hal-02296093⟩
102 Consultations
0 Téléchargements
Dernière date de mise à jour le 21/04/2024
comment ces indicateurs sont-ils produits

Altmetric

Partager

Gmail Facebook Twitter LinkedIn Plus