M step of the EM algorithm for fitting homogeneous Markov switching auto-regressive model swith constraints on the matrices, called in fit.MSAR. The matrices are constrained to have the same pattern ()zeros and non zeros coefficients) as the initial matrices.
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.
theta
model's parameter; object of class MSAR. See also init.theta.MSAR.
FB
Forward-Backward results, obtained by calling Estep.MSAR function
sigma.diag
if TRUE the innovation covariance matrices are diagonal.
Value
A list containing
A0
intercepts
A
AR coefficients
sigma
variance of innovation
prior
prior probabilities
transmat
transition matrix
References
Ailliot P., Monbet V., (2012), Markov switching autoregressive models for wind time series. Environmental Modelling & Software, 30, pp 92-101.