randindx {RIPSeeker} | R Documentation |
Returns an array of T indexes distributed as specified by p (which should be a normalized probability vector).
randindx(p, Total, NO_CHK)
p |
A row vector of normalized probabilities that dictate the transition probability from the current state to the next state. For example, p = [0.2, 0.8] indicates that the current state transitoins to state 1 at 0.2 and 2 at 0.8. The current state itself can either be the state 1 or 2. |
Total |
Total number of states needed to be generated using the input transition vector. |
NO_CHK |
Check whether the first argument is a valid row vector of normalized probabilities. |
The function is used by nbh_gen
to generate random data point based on the user-supplied transition probability matrix.
I |
Index/Indices or state(s) sampled following the transition. |
probability.
Yue Li
Capp\'e, O. (2001). H2M : A set of MATLAB/OCTAVE functions for the EM estimation of mixtures and hidden Markov models. (http://perso.telecom-paristech.fr/cappe/h2m/)
# Total contains the length of data to simulate Total <- 100 # number of states N <- 2 # transition probabilities between states TRANS <- matrix(c(0.9, 0.1, 0.3, 0.7), nrow=2, byrow=TRUE) label <- matrix(0, Total, 1) # Simulate initial state label[1] <- randindx(matrix(1,ncol=N)/N, 1, 1) # Use Markov property for the following time index for(t in 2:Total) { label[t] <- randindx(TRANS[label[t-1],], 1, 1) } plot(label)