inmar_hmm_mle: Maximum likelihood estimation of multivariate normal...

In longjess/hornsharkHMM: Functions For Analysing Horn Shark Data Using Hidden Markov Models

Maximum likelihood estimation of multivariate normal parameters

Description

Maximum likelihood estimation of multivariate normal parameters

Usage

inmar_hmm_mle(
  x,
  m,
  q,
  k,
  mu0,
  sigma0,
  gamma0,
  phi0,
  delta0 = NULL,
  stationary = TRUE,
  hessian = FALSE,
  steptol = 1e-06,
  iterlim = 100,
  stepmax = 100,
  state = NULL
)

Arguments

x	Matrix of observations, rows represent each variable
m	Number of states
mu0	List of vectors of length m, initial values for means for white noise
sigma0	List of vectors of length m, initial values for standard deviations
gamma0	Initial values for ransition probabiilty matrix, size m x m
phi0	List of matrices of size k x q, initial values for autoregressive parameters.
delta0	Optional, vector of length m containing initial values initial distribution
stationary	Boolean, whether the HMM is stationary or not
hessian	Boolean, whether to return the inverse hessian
state	List of state values, if provided. 0 represents an unknown state value.

Value

List of results

longjess/hornsharkHMM documentation built on June 15, 2022, 11:32 p.m.