A Comparison of MCMC Samplers for
Estimating Leveraged Stochastic Volatility
Model
Didit B. Nugroho1,2 and Takayuki Morimoto1
1Department of Mathematical Sciences, Kwansei Gakuin University, Japan 2Department of Mathematics, Satya Wacana Christian University, Indonesia
December 23, 2013, Hiroshima University of Economics
Abstract
1. Introduction: SV and MCMC methods
Stochastic volatility (SV) model is among the most important tools for modeling volatility of a financial time series.
A problem in parametric SV models is that it is not possible to obtain an explicit expression for the likelihood function of some unknown parameters.
An approach has become very attractive is Markov Chain Monte Carlo (MCMC) method proposed by Shephard (1993) and Jacquier et al (1994).
Omori and Watanabe (2008)proposed an efficient multi-move (block) Metropolis-Hastings sampler for sampling high-dimensional latent volatil-ity in the leveraged SV model.
2. Survey on MCMC Comparison
Kim et al. (1998) andJacquier-Polson (2011)compared the perfor-mance of the single-move and multi-move MH (MM-MH) samplers to estimate basic SV model. The MM-MH sampler has been proposed to reduce sample autocorrelations effectively. Jacquier-Polson par-ticularly note that the single-move and multi-move samplers deliver almost the same output.
Takaishi (2009) compared the performance of the single-move MH and HMC samplers and concluded that HMC sampler is superior to the single-move sampler.
3. Our Purpose
First purpose of this study is to extend the HMC and RMHMC sam-pling procedures proposed byGirolami and Calderhead (2011)for the leveraged SV model.
4. HMC and RMHMC Samplers
HMC-based methods are based on Hamiltonian dynamics system:
H(θ,ω) =−L(θ) +1
2log
n
(2π)D|M|o+12ω′M−1ω,
where L(θ) is the logarithm of the joint probability distribution for
the parameters θ∈RD,Mis the covariance matrix, andω∈RDis
the independent auxiliary variable.
In the RMHMC sampling,Mdepends on the variableθand is chosen
to be the metric tensor, i.e.
4. HMC and RMHMC Samplers (Cont’ed)
The full algorithm for HMC or RMHMC can then be summarized in the following three steps.
(1) Randomly draw a sample momentum vector ω∼ N(ω|0,M). (2) Run the leapfrog algorithm for NL steps with step size∆τ to
generate a proposal (θ∗,ω∗)according to the Hamiltonian equations
dθ
dτ = ∂H ∂ω and
dω
dτ =− ∂H
∂θ.
over a fictitious time τ.
(3) Accept (θ∗,ω∗)with probability
5. Leveraged SV Model
Omori and Watanabe (2008) proposed a leveraged SV model with normal errors (LSV-N model, hereafter) formulated as
Rt = σre
using MH algorithm separately and the latent volatilityhtusing
6. Comparison on Real Data
6. Comparison on Real Data (Cont’ed)
Tabel: Tuning parameters for the HMC and RMHMC implementations.
Parameter of model Sampler Parameter of
sampler h φ (σr,σh,ρ)
HMC N∆L 100 100 100
τ 0.01 0.0125 0.0125
RMHMC
NL 50 6 6
∆τ 0.1 0.5 0.5
NFPI - 5 5
6. Comparison on Real Data (Cont’ed)
Tabel: Posterior summary statistics.
Parameter Mean (SD) 95%Interval IACT Time (sec) Panel A: Using MM-MH sampler
σr 1.259(0.070) [1.121, 1.398] 20.8
−
φ 0.945(0.019) [0.902, 0.974] 118.2
σh 0.193(0.033) [0.138, 0.267] 206.7
ρ −0.442(0.103) [−0.630,−0.231] 92.7 Panel B: Using HMC sampler
σr 1.241(0.070) [1.098, 1.377] 53.0
465.88
φ 0.954(0.014) [0.925, 0.980] 132.7
σh 0.169(0.025) [0.120, 0.219] 279.9
ρ −0.461(0.097) [−0.648,−0.272] 83.9 Panel C: Using RMHMC sampler
σr 1.238(0.070) [1.098, 1.378] 19.4
6. Comparison on Real Data (Cont’ed)
Tabel: Acceptance rates.
Acceptance rates Parameter
MM-MH HMC RMHMC
h 0.856 0.889 0.975
φ 0.955 0.910 0.901
7. Conclusion
Based on the empirical results, we conclude:
The RMHMC sampler give the best performance in terms of autocor-relation time.
In particular, the HMC sampler exhibits slightly slower convergence than MM-HMC sampler except for the leverage parameter.
Some References
Girolami, M. and Calderhead, B.(2011). Riemann manifold Langevin and
Hamiltonian Monte Carlo methods.Journal of the Royal Statistical Society, Series B,73(2), 1–37.
Jacquier, E. and Polson, N. G. (2011). Bayesian methods in finance. In J.
Geweke, G. Koop, and H. van Dijk (Eds.),Handbook of Bayesian Econometrics. Oxford University Press.
Kim, S. and Shephard, N. and Chib, S.(1998). Stochastic volatility: likelihood
inference and comparison with ARCH models. In N. Shephard (Ed.),Stochastic Volatility: Selected Readings. Oxford University Press.
Omori, Y. and Watanabe, T.(2008). Block sampler and posterior mode
esti-mation for asymmetric stochastic volatility models.Computational Statistics and Data Analysis,52, 2892–2910.