Functions that compute the initial probability distribution for the states. Currently only CJS, MS models and MS models with state uncertainty are included and these all use cjs_delta to assign a known state.
cjs_delta(pars, m, F, T, start)
Arguments
pars: list of real parameter matrices (id by occasion) for each type of parameter
m: number of states
F: initial occasion vector
T: number of occasions
start: matrix with values that are first occasion and for some CJS type models the state of first observation
Returns
2-d array of initial state probability vectors for each id
References
Zucchini, W. and I.L. MacDonald. 2009. Hidden Markov Models for Time Series: An Introduction using R. Chapman and Hall, Boca Raton, FL. 275p.