sampleCommonLogVols: Sample the latent log-volatilities, common to m-dimensional...

In drkowal/FDLM: A Bayesian Multivariate Functional Dynamic Linear Model

Description

Compute one draw of the log-volatilities using a discrete mixture of Gaussians approximation to the likelihood (see Omori, Chib, Shephard, and Nakajima, 2007) where the log-vols are assumed to follow an AR(1) model. The model assumes that the volatility are common to an m-dimensional time series.

Usage

sampleCommonLogVols(h_y, h_prev, h_mu, h_phi, h_sigma_eta)

Arguments

h_y	the T x m matrix of data, which follows one join SV model
h_prev	the T x 1 matrix of the previous log-vols
h_mu	the log-vol unconditional mean
h_phi	the log-vol AR(1) coefficient
h_sigma_eta	the log-vol innovation standard deviation

Value

T x 1 matrix of simulated log-vols

drkowal/FDLM documentation built on May 20, 2019, 5:20 p.m.