Mstep.classif: fit an AR model for each class of C

In NHMSAR: Non-Homogeneous Markov Switching Autoregressive Models

Description

fit an AR model for each class of C by maximum likelihood method.

Usage

Mstep.classif(data, C, order,sigma.diag=FALSE)

Arguments

data	array of univariate or multivariate series with dimension T*N.samples*d. T: number of time steps of each sample, N.samples: number of realisations of the same stationary process, d: dimension.
C	Class sequence
order	order of AR models (all models will have the same order)
sigma.diag	if TRUE the covariance matrices will be diagonal (default FALSE)

Value

list containing

A0	intercept
A	AR coefficients
sigma	variance of innovation
LL	log likelihood

Author(s)

Valerie Monbet, [e-mail]

See Also

fit.MSAR

Examples

	data(meteo.data)
data = array(meteo.data$temperature,c(31,41,1)) 
k = 40
T = dim(data)[1]
N.samples = dim(data)[2]
d = dim(data)[3]
order = 2
C = array(meteo.data>0,c(31,41,1))
res = Mstep.classif(data,C,order=order)
str(res)

NHMSAR documentation built on Feb. 9, 2022, 9:06 a.m.