InformationMatrixARp {FitAR}R Documentation

Information Matrix for ARp

Description

The large-sample information matrix per observation is computed in a subset AR with the usual parameterization, that is, a subset of the AR coefficients.

Usage

InformationMatrixARp(phi, lags)

Arguments

phi vector of coefficients in the subset AR
lags vector indicating lags present in phi

Details

The subset information matrix is obtained simply by selecting the appropriate rows and columns from the full information matrix. This function is used by FitARLS to obtain the estimated standard errors of the parameter estimates.

Value

a p-by-p Toeplitz matrix, p=length(phi)

Author(s)

A.I. McLeod

References

McLeod, A.I. and Zhang, Y. (2006). Partial autocorrelation parameterization for subset autoregression. Journal of Time Series Analysis, 27, 599-612.

See Also

InformationMatrixAR, FitARLS, InformationMatrixARz

Examples

#variances of parameters in a subset ARp(1,2,6)
fi<-InformationMatrixARp(c(0.36,0.23,0.23),c(1,2,6))
sqrt(diag(solve(fi*197)))

[Package FitAR version 1.0 Index]