HMMSim {RHmm} | R Documentation |
Simulation of an HMM for different classes of observations distributions
HMMSim(nSim, HMM, lastState=NULL)
nSim |
Number of simulations |
HMM |
An HMMClass object. See HMMSet |
lastState |
If not NULL, the state of the previous observation (usefull to complete a serie) |
a list with
obs |
simulated observations (a vector for univariate distributions, a matrix for multivariate distributions) |
states |
simulated hidden states |
codeHMMSet
# simulate a 3 hidden states model with univariate normal distributions n_1d_3s <- distributionSet("NORMAL", mean=c(1, -2, 5), var=c(1, 2, 4)) initProb3 <- rep(1,3)/3 transMat3 <- rbind(c(0.5, 0.4, 0.1), c(0.3, 0.4, 0.3), c(0.2, 0.1, 0.7)) hmm_1d_3s <- HMMSet(initProb3, transMat3, n_1d_3s) simul <- HMMSim(1000, hmm_1d_3s) #Complete the serie. simul <- c(simul, HMMSim(1000, hmm_1d_3s, simul$states[1000]))