armaStatistics {fSeries}R Documentation

Statisitcs of the True ARMA Process

Description

A collection of functions for the analysis of the true ARMA process. The first investigates the characteristic polynomial and the second the true autocorrelation function.

The functions are:

1 armaRoots Roots of the characteristic ARMA polynomial,
2 armaTrueacf True autocorrelation function of an ARMA process.

Usage

armaRoots(coefficients, n.plot = 400, digits = 4, ...) 
armaTrueacf(model, lag.max = 20, type = "correlation", doplot = TRUE) 

Arguments

coefficients a numeric vector with the coefficients of the characterisitic polynomial.
digits output precision, an integer value.
doplot a logical. Should a plot be displayed?
lag.max maximum number of lags at which to calculate the acf or pacf, an integer value by default 20.
model a specification of the ARMA model with two elements: model$ar is the vector of the AR coefficients, and model$ma is the vector of the MA coefficients.
n.plot the number of data point to plot the unit circle; an integer value.
type a character string, "correlation" to compute the true autocorrelation function, "partial" to compute the true partial autocorrelation function, or "both" if both functions are desired. The start of one of the strings will suffice.
... arguments past to the plot function.

Value

armaRoots
returns a three column data frame with the real, the imaginary part and the radius of the roots. The number of rows corresponds to the coefficients.
armaTrueacf
returns a two column data frame with the lag and the correlation function.

Author(s)

B.D. Ripley for the underlying ARMAacf function,
Diethelm Wuertz for this R-port.

See Also

armaModelling.

Examples

## armaRoots -
   xmpSeries("\nStart: Roots of the Characteristic Polynomial > ")
   # Calculate and plot the  of an ARMA process:
   par(mfrow = c(2, 2), cex = 0.7)
   coefficients = c(-0.5, 0.9, -0.1, -0.5)
   armaRoots(coefficients)
   
## armaTrueacf -
   xmpSeries("\nNext: True ACF of an ARMA Process > ")
   model = list(ar = c(0.3, +0.3), ma = 0.1)
   armaTrueacf(model)
   model = list(ar = c(0.3, -0.3), ma = 0.1)
   armaTrueacf(model)

[Package Contents]