FCIplot: FCIplot

In Eplot: Plotting longitudinal series

Description

Estimate and plot prediction standard deviation. Given the series, the function estimate point prediction based on AR(1) model and, using the resdiuals from this simple model, estimate an ARCH model to estimate the prediction standard deviation. If plott=TRUE, a plot of the most recent k values is created.

Usage

FCIplot(series, plott = TRUE, wind1 = 24, wind2 = 60, k = 60,
  rrr1 = "Rec", rrr2 = "Rec", main = "series")

Arguments

series	series to be plotted.
plott	should a plot be created? default is plott=TRUE.
wind1	window size for the AR component (see details).
wind2	window size for the ARCH component (see details).
k	if plott=TRUE, tail(series,k) will be plotted.
rrr1	will the AR model be estimated using Recursive ("Rec") or Rolling ("Rol") window?
rrr2	will the ARCH model be estimated using Recursive ("Rec") or Rolling ("Rol") window?
main	main title of the plot, same as in plot.default.

Details

Estimate and plot prediction confidence intervals based on AR-ARCH model.

Value

vector of prediction's standard deviation.

Examples

par(mfrow = c(2,1))
out <- FCIplot(rnorm(100),plott=TRUE,k=30)
plott(out,main="The out-of-sample standard deviation")

Eplot documentation built on May 2, 2019, 2:42 a.m.