pomp (version 2.6)

pfilter: Particle filter

Description

A plain vanilla sequential Monte Carlo (particle filter) algorithm. Resampling is performed at each observation.

Usage

# S4 method for data.frame
pfilter(
  data,
  Np,
  tol = 1e-17,
  max.fail = Inf,
  params,
  rinit,
  rprocess,
  dmeasure,
  pred.mean = FALSE,
  pred.var = FALSE,
  filter.mean = FALSE,
  filter.traj = FALSE,
  save.states = FALSE,
  ...,
  verbose = getOption("verbose", FALSE)
)
# S4 method for pomp
pfilter(
  data,
  Np,
  tol = 1e-17,
  max.fail = Inf,
  pred.mean = FALSE,
  pred.var = FALSE,
  filter.mean = FALSE,
  filter.traj = FALSE,
  save.states = FALSE,
  ...,
  verbose = getOption("verbose", FALSE)
)
# S4 method for pfilterd_pomp
pfilter(data, Np, tol, ..., verbose = getOption("verbose", FALSE))
# S4 method for objfun
pfilter(data, ...)

Arguments

data

either a data frame holding the time series data, or an object of class ‘pomp’, i.e., the output of another pomp calculation.

Np

the number of particles to use. This may be specified as a single positive integer, in which case the same number of particles will be used at each timestep. Alternatively, if one wishes the number of particles to vary across timesteps, one may specify Np either as a vector of positive integers of length

length(time(object,t0=TRUE))

or as a function taking a positive integer argument. In the latter case, Np(k) must be a single positive integer, representing the number of particles to be used at the k-th timestep: Np(0) is the number of particles to use going from timezero(object) to time(object)[1], Np(1), from timezero(object) to time(object)[1], and so on, while when T=length(time(object,t0=TRUE)), Np(T) is the number of particles to sample at the end of the time-series.

tol

non-negative numeric scalar; particles with likelihood less than tol are considered to be incompatible with the data. See the section on Filtering Failures for more information. In a future release, this argument will be removed.

max.fail

integer; the maximum number of filtering failures allowed (see below). If the number of filtering failures exceeds this number, execution will terminate with an error. By default, max.fail is set to infinity, so no error can be triggered. In a future release, this argument will be removed.

params

optional; named numeric vector of parameters. This will be coerced internally to storage mode double.

rinit

simulator of the initial-state distribution. This can be furnished either as a C snippet, an R function, or the name of a pre-compiled native routine available in a dynamically loaded library. Setting rinit=NULL sets the initial-state simulator to its default. For more information, see ?rinit_spec.

rprocess

simulator of the latent state process, specified using one of the rprocess plugins. Setting rprocess=NULL removes the latent-state simulator. For more information, see ?rprocess_spec for the documentation on these plugins.

dmeasure

evaluator of the measurement model density, specified either as a C snippet, an R function, or the name of a pre-compiled native routine available in a dynamically loaded library. Setting dmeasure=NULL removes the measurement density evaluator. For more information, see ?dmeasure_spec.

pred.mean

logical; if TRUE, the prediction means are calculated for the state variables and parameters.

pred.var

logical; if TRUE, the prediction variances are calculated for the state variables and parameters.

filter.mean

logical; if TRUE, the filtering means are calculated for the state variables and parameters.

filter.traj

logical; if TRUE, a filtered trajectory is returned for the state variables and parameters. See filter.traj for more information.

save.states

logical. If save.states=TRUE, the state-vector for each particle at each time is saved.

...

additional arguments supply new or modify existing model characteristics or components. See pomp for a full list of recognized arguments.

When named arguments not recognized by pomp are provided, these are made available to all basic components via the so-called userdata facility. This allows the user to pass information to the basic components outside of the usual routes of covariates (covar) and model parameters (params). See ?userdata for information on how to use this facility.

verbose

logical; if TRUE, diagnostic messages will be printed to the console.

Value

An object of class ‘pfilterd_pomp’, which extends class ‘pomp’.

Methods

logLik

the estimated log likelihood

cond.logLik

the estimated conditional log likelihood

eff.sample.size

the (time-dependent) estimated effective sample size

pred.mean, pred.var

the mean and variance of the approximate prediction distribution

filter.mean

the mean of the filtering distribution

filter.traj

retrieve one particle trajectory. Useful for building up the smoothing distribution.

as.data.frame

coerce to a data frame

plot

diagnostic plots

Filtering failures

If the degree of disagreement between model and data becomes sufficiently large, a “filtering failure” results. A filtering failure occurs when, at some time point, none of the Np particles is compatible with the data. In particular, if the conditional likelihood of a particle at any time is below the tolerance value tol, then that particle is considered to be uninformative and its likelihood is taken to be zero. A filtering failure occurs when this is the case for all particles. A warning is generated when this occurs unless the cumulative number of failures exceeds max.fail, in which case an error is generated.

References

M.S. Arulampalam, S. Maskell, N. Gordon, & T. Clapp. A Tutorial on Particle Filters for Online Nonlinear, Non-Gaussian Bayesian Tracking. IEEE Transactions on Signal Processing 50, 174--188, 2002.

See Also

Other elementary POMP methods: pomp-package, probe(), simulate(), spect()

Other particle filter methods: bsmc2(), cond.logLik(), eff.sample.size(), filter.mean(), filter.traj(), mif2(), pmcmc(), pred.mean(), pred.var()

Examples

# NOT RUN {
pf <- pfilter(gompertz(),Np=1000)	## use 1000 particles

plot(pf)
logLik(pf)
cond.logLik(pf)			## conditional log-likelihoods
eff.sample.size(pf)             ## effective sample size
logLik(pfilter(pf))      	## run it again with 1000 particles

## run it again with 2000 particles
pf <- pfilter(pf,Np=2000,filter.mean=TRUE,filter.traj=TRUE)
fm <- filter.mean(pf)    	## extract the filtering means
ft <- filter.traj(pf)    	## one draw from the smoothing distribution
# }