smooth.hmm: Calculate the probability of being in a particular state for...

In rarhsmm: Regularized Autoregressive Hidden Semi Markov Model

Description

Calculate the probability of being in a particular state for each observation.

Usage

smooth.hmm(y, mod)

Arguments

y

observed series

mod

list consisting the at least the following items: mod$m = scalar number of states, mod$delta = vector of initial values for prior probabilities, mod$gamma = matrix of initial values for state transition probabilies. mod$mu = list of initial values for means, mod$sigma = list of initial values for covariance matrices. For autoregressive hidden markov models, we also need the additional items: mod$arp = scalar order of autoregressive structure mod$auto = list of initial values for autoregressive coefficient matrices

Value

a matrix containing the state probabilities

References

Rabiner, Lawrence R. "A tutorial on hidden Markov models and selected applications in speech recognition." Proceedings of the IEEE 77.2 (1989): 257-286.

Examples

set.seed(15562)
m <- 2
mu <- list(c(3,4,5),c(-2,-3,-4))
sigma <- list(diag(1.3,3), 
            matrix(c(1,-0.3,0.2,-0.3,1.5,0.3,0.2,0.3,2),3,3,byrow=TRUE))
delta <- c(0.5,0.5)
gamma <- matrix(c(0.8,0.2,0.1,0.9),2,2,byrow=TRUE)
auto <- list(matrix(c(0.3,0.2,0.1,0.4,0.3,0.2,
                     -0.3,-0.2,-0.1,0.3,0.2,0.1,
                      0,0,0,0,0,0),3,6,byrow=TRUE),
            matrix(c(0.2,0,0,0.4,0,0,
                      0,0.2,0,0,0.4,0,
                     0,0,0.2,0,0,0.4),3,6,byrow=TRUE))
mod <- list(m=m,mu=mu,sigma=sigma,delta=delta,gamma=gamma,auto=auto,arp=2)
sim <- hmm.sim(2000,mod)
y <- sim$series
state <- sim$state
fit <- em.hmm(y=y, mod=mod, arp=2)
stateprob <- smooth.hmm(y=y,mod=fit)
head(cbind(state,stateprob),20)

Example output

iteration  1 ; loglik =  -10052.34 
iteration  2 ; loglik =  -10014.45 
iteration  3 ; loglik =  -10014.45 
      state                           
 [1,]     1 1.000000e+00 3.211755e-139
 [2,]     1 1.000000e+00  4.256785e-40
 [3,]     1 1.000000e+00  6.211848e-37
 [4,]     1 1.000000e+00  3.070092e-40
 [5,]     1 1.000000e+00  1.890428e-25
 [6,]     2 2.533539e-34  1.000000e+00
 [7,]     2 1.300674e-17  1.000000e+00
 [8,]     2 1.202568e-20  1.000000e+00
 [9,]     1 1.000000e+00  1.227771e-22
[10,]     1 1.000000e+00  2.649681e-23
[11,]     1 1.000000e+00  1.640768e-33
[12,]     1 1.000000e+00  1.252117e-32
[13,]     1 1.000000e+00  6.673570e-36
[14,]     1 1.000000e+00  7.154855e-36
[15,]     1 1.000000e+00  9.100629e-34
[16,]     2 8.020779e-32  1.000000e+00
[17,]     1 1.000000e+00  3.188145e-19
[18,]     1 1.000000e+00  3.385804e-26
[19,]     1 1.000000e+00  1.537049e-31
[20,]     1 1.000000e+00  3.535321e-33

rarhsmm documentation built on May 2, 2019, 9:33 a.m.