hsmmviterbi_exp: Viterbi algorithm to decode the latent states in hidden...

In ziphsmm: Zero-Inflated Poisson Hidden (Semi-)Markov Models

Description

Viterbi algorithm to decode the latent states in hidden semi-Markov models with covariates where the latent state durations have accelerated failure time structure

Usage

hsmmviterbi_exp(y, M, trunc, dtrate, dtparm, prior, zeroparm, emitparm, tpmparm,
  dt_x, zeroinfl_x, emit_x, tpm_x, plot = FALSE, xlim = NULL, ylim = NULL,
  ...)

Arguments

y	the observed series to be decoded
M	number of latent states
trunc	a vector specifying the truncation at the maximum number of dwelling time in each state.
dtrate	a vector for the scale parameters in the base exponential density for the latent state durations.
dtparm	a matrix of coefficients for the accelerated failure time model in each latent state
prior	a vector of prior probabilities
zeroparm	a vector of regression coefficients for the structural zero proportion in state 1
emitparm	a matrix of regression coefficients for the Poisson regression in each state
tpmparm	a vector of coefficients for the multinomial logistic regression in the transition probabilities
dt_x	a matrix of covariates for the latent state durations
zeroinfl_x	a matrix of covariates for the zero proportion
emit_x	a matrix of covariates for the Poisson means
tpm_x	a matrix of covariates for the transition
plot	whether a plot should be returned
xlim	vector specifying the minimum and maximum on the x-axis in the plot. Default to NULL.
ylim	vector specifying the minimum and maximum on the y-axis in the plot. Default to NULL.
...	further arguments to be passed to the plot() function

Value

decoded series of latent states

References

Walter Zucchini, Iain L. MacDonald, Roland Langrock. Hidden Markov Models for Time Series: An Introduction Using R, Second Edition. Chapman & Hall/CRC

Examples

## Not run: 
M <- 3
prior <- c(0.5,0.3,0.2)
dtrate <- c(6,5,4)
dtparm <- matrix(c(0.2,0.1,0.2),nrow=3)
zeroparm <- c(0,-0.2)
emitparm <- matrix(c(4,0.3,5,0.2,6,-0.1),3,2,byrow=TRUE)
tpmparm <- c(1,0.2,0.5,-0.2,0,0.2)

emit_x <- matrix(c(rep(1,1000),rep(0,1000)),nrow=2000,ncol=1)
dt_x <- emit_x
tpm_x <- emit_x
zeroinfl_x <- emit_x
trunc <- c(18,15,10)

re <- hsmmsim2_exp(prior,dtrate,dtparm,zeroparm,emitparm,tpmparm,
                  trunc, M, n, dt_x,tpm_x, emit_x, zeroinfl_x)
y <- re$series

rrr <- hsmmfit_exp(y,M,trunc,dtrate,dtparm,prior,zeroparm,emitparm,tpmparm,
                  dt_x,zeroinfl_x,emit_x,tpm_x,method="BFGS",control=list(trace=1))

decode <- hsmmviterbi_exp(y,M, trunc,dtrate,dtparm,
                          prior,zeroparm,emitparm,tpmparm,
                          dt_x, zeroinfl_x, emit_x, tpm_x)
sum(decode!=re$state)

## End(Not run)

ziphsmm documentation built on May 2, 2019, 6:10 a.m.