rprocess: rprocess

Description Usage Arguments Details Value See Also

Description

rprocess simulates the process-model portion of partially-observed Markov process.

Usage

1
2
3
## S4 method for signature 'pomp'
rprocess(object, xstart, times, params, offset = 0L,
  ...)

Arguments

object

an object of class ‘pomp’, or of a class that extends ‘pomp’. This will typically be the output of pomp, simulate, or one of the pomp inference algorithms.

xstart

an nvar x nrep matrix containing the starting state of the system. Columns of xstart correspond to states; rows to components of the state vector. One independent simulation will be performed for each column. Note that in this case, params must also have nrep columns.

times

a numeric vector (length ntimes) containing times. These must be in non-decreasing order.

params

a npar x nrep matrix of parameters. Each column is treated as an independent parameter set, in correspondence with the corresponding column of xstart.

offset

integer; the first offset times in times will be discarded.

...

additional arguments are ignored.

Details

When rprocess is called, the first entry of times is taken to be the initial time (i.e., that corresponding to xstart). Subsequent times are the additional times at which the state of the simulated processes are required.

Value

rprocess returns a rank-3 array with rownames. Suppose x is the array returned. Then

1
dim(x)=c(nvars,nrep,ntimes-offset),

where nvars is the number of state variables (=nrow(xstart)), nrep is the number of independent realizations simulated (=ncol(xstart)), and ntimes is the length of the vector times. x[,j,k] is the value of the state process in the j-th realization at time times[k+offset]. The rownames of x must correspond to those of xstart.

See Also

Specification of the process-model simulator: rprocess_spec

Other pomp workhorses: dmeasure, dprior, dprocess, partrans, rinit, rmeasure, rprior, skeleton, workhorses


kidusasfaw/pomp documentation built on May 20, 2019, 2:59 p.m.