NHMSAR (version 1.0)

Mstep.nn.MSAR: M step of the EM algorithm.

Description

M step of the EM algorithm for fitting Markov switching auto-regressive models with non homogeneous emissions and non homogeneous transitions.

Usage

Mstep.nn.MSAR(data, theta, FB, 
   covar.trans = covar.trans, covar.emis = covar.emis, method = NULL)

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.

theta

model's parameter; object of class MSAR. See also init.theta.MSAR.

FB

Forward-Backward results, obtained by calling Estep.MSAR function

covar.trans

transitions covariates

covar.emis

emissions covariates (the covariates act on the intercepts)

method

permits to choice the optimization algorithm. default is "ucminf", other possible choices are "BFGS" or "L-BFGS-B

Value

• A0
  intercepts
• A
  AR coefficients
• sigma
  variance of innovation
• prior
  prior probabilities
• transmat
  transition matrix
• par_emis
  emission parameters
• par.trans
  transitions parameters

References

Ailliot P., Monbet V., (2012), Markov switching autoregressive models for wind time series. Environmental Modelling & Software, 30, pp 92-101.

See Also

Mstep.hh.MSAR