2019/02/25 by Chiara Lattanzi, Lattanzi, Chiara, Manuele Leonelli +1 · 1 citation
Economics, Econometrics and Finance · #FOS: Computer and information sciences #FOS: Economics and business #Financial Risk and Volatility Modeling #Market Dynamics and Volatility #Methodology (stat.ME) #Monetary Policy and Economic Impact #Statistical Finance (q-fin.ST)
paper · pdf · doi:10.48550/arxiv.1902.09205
openalex publication_date 2019/02/25 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28
Inference over tails is usually performed by fitting an appropriate limiting\ndistribution over observations that exceed a fixed threshold. However, the\nchoice of such threshold is critical and can affect the inferential results.\nExtreme value mixture models have been defined to estimate the threshold using\nthe full dataset and to give accurate tail estimates. Such models assume that\nthe tail behavior is constant for all observations. However, the extreme\nbehavior of financial returns often changes considerably in time and such\nchanges occur by sudden shocks of the market. Here we extend the extreme value\nmixture model class to formally take into account distributional extreme\nchangepoints, by allowing for the presence of regime-dependent parameters\nmodelling the tail of the distribution. This extension formally uses the full\ndataset to both estimate the thresholds and the extreme changepoint locations,\ngiving uncertainty measures for both quantities. Estimation of functions of\ninterest in extreme value analyses is performed via MCMC algorithms. Our\napproach is evaluated through a series of simulations, applied to real data\nsets and assessed against competing approaches. Evidence demonstrates that the\ninclusion of different extreme regimes outperforms both static and dynamic\ncompeting approaches in financial applications.\n