作者: Nima Nonejad
DOI:
关键词:
摘要: This paper proposes a model that simultaneously captures long memory and structural breaks. We breaks through irreversible Markov switching or so-called change-point dynamics. The parameters subject to the unobserved states which determine position of are sampled from joint posterior density by sampling their respective conditional posteriors using Gibbs Metropolis-Hastings. Monte Carlo simulations demonstrate proposed estimation approach is effective in identifying dating Applied daily S&P 500 data, one finds strong evidence three these robust different specifications including GARCH specification for variance volatility.