covmat.fsvdraws: Extract posterior draws of the model-implied covariance...

In factorstochvol: Bayesian Estimation of (Sparse) Latent Factor Stochastic Volatility Models

Extract posterior draws of the model-implied covariance matrix

Description

covmat extracts draws from the model-implied covariance matrix from an fsvdraws object for all points in time which have been stored.

Usage

## S3 method for class 'fsvdraws'
covmat(x, timepoints = "all", ...)

Arguments

x	Object of class 'fsvdraws', usually resulting from a call of fsvsample.
timepoints	Vector indicating at which point(s) in time (of those that have been stored during sampling) the correlation matrices should be extracted. Can also be "all" or "last".
...	Ignored.

Value

Array of dimension m times m times draws times timepoints containing the posterior draws for the model-implied covariance matrix.

Note

Currently crudely implemented as a double loop in pure R, may be slow.

See Also

Other extractors: cormat.fsvdraws(), runningcormat(), runningcovmat()

Examples

set.seed(1)
sim <- fsvsim(n = 500, series = 3, factors = 1) # simulate
res <- fsvsample(sim$y, factors = 1, keeptime = "all") # estimate
covs <- covmat(res, "last") # extract

# Trace plot of determinant of posterior covariance matrix
# at time t = n = 500:
detdraws <- apply(covs[,,,1], 3, det)
ts.plot(detdraws)
abline(h = mean(detdraws), col = 2)          # posterior mean
abline(h = median(detdraws), col = 4)        # posterior median
abline(h = det(covmat(sim, "last")[,,1]), col = 3) # implied by DGP

# Trace plot of draws from posterior covariance of Sim1 and Sim2 at
# time t = n = 500:
ts.plot(covs[1,2,,1])
abline(h = covmat(sim, "last")[1,2,1], col = 3) # "true" value

# Smoothed kernel density estimate:
plot(density(covs[1,2,,1], adjust = 2))

# Summary statistics:
summary(covs[1,2,,1])

factorstochvol documentation built on Nov. 24, 2023, 5:08 p.m.