Eplot (version 1.0)

FCIplot: FCIplot

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.

Value

vector of prediction's standard deviation.

Details

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

Examples

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