Extracting volatility signal using maximum a posteriori estimation
2016
This paper outlines a methodology to estimate a denoised volatility signal for foreign exchange rates using a hidden Markov model (HMM). For this purpose a maximum a posteriori (MAP) estimation is performed. A double exponential prior is used for the state variable (the log-volatility) in order to allow sharp jumps in realizations and then log-returns marginal distributions with heavy tails. We consider two routes to choose the regularization and we compare our MAP estimate to realized volatility measure for three exchange rates.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
16
References
0
Citations
NaN
KQI