Predicts correlation matrix
predcor simulates from the posterior predictive distribution
of the model-implied correlation matrix.
predcor(x, ahead = 1, each = 1)
x |
Object of class |
ahead |
Vector of timepoints, indicating how many steps to predict ahead. |
each |
Single integer (or coercible to such) indicating how often should be drawn from the posterior predictive distribution for each draw that has been stored during MCMC sampling. |
4-dimensional array containing draws from the predictive correlation distribution.
Currently crudely implemented as a triple loop in pure R, may be slow.
Other predictors:
predcond(),
predcov(),
predh(),
predloglikWB(),
predloglik(),
predprecWB()
set.seed(1) sim <- fsvsim(series = 3, factors = 1) # simulate res <- fsvsample(sim$y, factors = 1) # estimate # Predict 1, 10, and 100 days ahead: predobj <- predcor(res, ahead = c(1, 10, 100)) # Trace plot of draws from posterior predictive distribution # of the correlation of Sim1 and Sim2: # (one, ten, and 100 days ahead): plot.ts(predobj[1,2,,]) # Smoothed kernel density estimates of predicted covariance # of Sim1 and Sim2: plot(density(predobj[1,2,,"1"], adjust = 2)) lines(density(predobj[1,2,,"10"], adjust = 2), col = 2) lines(density(predobj[1,2,,"100"], adjust = 2), col = 3)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.