forc.ecdf: Empirical CDF computations for posterior forecast samples
In MSBVAR: Markov-Switching, Bayesian, Vector Autoregression Models

Description Usage Arguments Details Value Author(s)

Computes (pointwise over time) empirical density (error bands) and mean forecasts for a Monte Carlo or Bayesian posterior sample of forecasts.

1	forc.ecdf(forecasts, probs = c(0.05, 0.95), start = c(0, 1), ...)

`forecasts`	Posterior sample of VAR forecasts produced by `hc.forecast.VAR()` or `uc.forecast.VAR()`
`probs`	Error band width in percentiles, default is 90% error band.
`start`	Start value for the time series – as in the `ts()` for the forecast horizon
`...`	Other `ecdf()` parameters

For each endogenous variable in the VAR and each point in the forecast horizon this function estimates the percentile based confidence interval. It then returns a time series matrix beginning at start of the mean forecast and the limits of the confidence region for each variable in the forecast sample.

A multiple time series object is returned where the first column is the mean estimate followed by the upper and lower bounds of the confidence region.

Patrick T. Brandt

MSBVAR documentation built on May 30, 2017, 1:23 a.m.