mixAR (version 0.22.8)

MixVARGaussian-class: MixVAR models with multivariate Gaussian noise components

Description

Class MixVARGaussian represents MixAR models with multivariate Gaussian noise components.

Objects from the Class

Objects can be created by calls of the form new("MixVARGaussian", ...), giving the elements of the model as named arguments, see the examples below.

arcoef contains the AR coefficients, one numeric array for each mixture component. It can be given as a "raggedCoefV" object or as a list of numeric arrays.

The AR order of the model is inferred from arcoef argument. If argument order is given, it is checked for consistency with arcoef. The shift slot defaults to a matrix of zeroes and the vcov slot to an array of identity matrices, one for each component.

The distribution of the noise components is standard multivariate Gaussian, N(0,1).

Slots

All slots except arcoef are numeric vectors of length equal to the number of components in the model.

prob:

probabilities of the mixture components,

order:

AR orders of the components,

shift:

the shift (intercept) terms of the AR components,

vcov:

covariance matrices of the noise terms of the AR components,

arcoef:

The AR components, object of class "raggedCoefV".

Extends

Class "MixAR", directly.

Methods

fit_mixAR

signature(x = "ANY", model = "MixARGaussian"): ...

Author

Davide Ravagli

See Also

class "MixAR"

Examples

showClass("MixVARGaussian")

## Create array of covariance matrices
Sigma1 <- cbind(c(0.0013, 0.0011), c(0.0011, 0.0012))
Sigma2 <- cbind(c(0.0072, 0.0047), c(0.0047, 0.0039))
Sigma  <- array(c(Sigma1, Sigma2), dim=c(2,2,2))

## Create list of AR coefficients
AR <- list()
AR[[1]] <- array(c(0.0973, -0.0499,  0.2927,  0.4256,  ## VAR(2;4)
                  -0.0429,  0.0229, -0.1515, -0.1795,
                  -0.0837, -0.1060, -0.1530,  0.1947,
                  -0.1690, -0.0903,  0.1959,  0.0955), dim=c(2,2,4))
AR[[2]] <- array(c(0.3243,  0.2648,  0.4956,  0.2870,  ## VAR(2;3)
                  -0.1488,  0.0454, -0.0593, -0.3629,
                   0.1314,  0.0274,  0.0637,  0.0485), dim=c(2,2,3))

## Create vector of mixing weights
prob <- c(0.6376, 0.3624)

## Create matrix of shift parameters
shift <- cbind(c(0.0044, 0.0020), c(-0.0039, -0.0014))

## Build "MixVARGaussian" model
new("MixVARGaussian", prob=prob, vcov=Sigma, arcoef=AR, shift=shift)