Trajectories drawn from the smoothing distribution
# S4 method for pfilterd_pomp
filter.traj(object, vars, ...)# S4 method for pfilterList
filter.traj(object, vars, ...)
# S4 method for pmcmcd_pomp
filter.traj(object, vars, ...)
# S4 method for pmcmcList
filter.traj(object, vars, ...)
result of a filtering computation
optional character; names of variables
ignored
The smoothing distribution is the distribution of $$X_t | Y_1=y^*_1, \dots, Y_T=y^*_T,$$ where \(X_t\) is the latent state process, \(Y_t\) is the observable process, \(t\) is time, and \(T\) is the time of the final observation.
In a particle filter, the trajectories of the individual particles are not independent of one another, since they share ancestry.
However, a randomly sampled particle trajectory \(X_1,\dots,X_T\) is a draw from the smoothing distribution.
Seting filter.traj = TRUE in pfilter causes one such trajectory to be sampled.
By running multiple independent pfilter operations, one can thus build up a picture of the smoothing distribution.
In particle MCMC (pmcmc), this operation is performed at each MCMC iteration.
Assuming the MCMC chain has converged, and after proper measures are taken to assure approximate independence of samples, filter.traj allows one to extract a sample from the smoothing distribution.
Other particle filter methods: bsmc2,
cond.logLik, eff.sample.size,
filter.mean, mif2,
pfilter, pmcmc,
pred.mean, pred.var