[Pomp-commits] r138 - pkg/man
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Tue Jun 2 21:20:27 CEST 2009
Author: kingaa
Date: 2009-06-02 21:20:25 +0200 (Tue, 02 Jun 2009)
New Revision: 138
Modified:
pkg/man/bsplines.Rd
pkg/man/dmeasure-pomp.Rd
pkg/man/dprocess-pomp.Rd
pkg/man/euler.Rd
pkg/man/euler.sir.Rd
pkg/man/eulermultinom.Rd
pkg/man/init.state-pomp.Rd
pkg/man/mif-class.Rd
pkg/man/mif-methods.Rd
pkg/man/mif.Rd
pkg/man/nlf.Rd
pkg/man/ou2.Rd
pkg/man/particles-mif.Rd
pkg/man/pfilter.Rd
pkg/man/pomp-class.Rd
pkg/man/pomp-methods.Rd
pkg/man/pomp-package.Rd
pkg/man/pomp.Rd
pkg/man/rmeasure-pomp.Rd
pkg/man/rprocess-pomp.Rd
pkg/man/simulate-pomp.Rd
pkg/man/skeleton-pomp.Rd
pkg/man/sobol.Rd
pkg/man/trajectory-pomp.Rd
Log:
further improvements to documentation
Modified: pkg/man/bsplines.Rd
===================================================================
--- pkg/man/bsplines.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/bsplines.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -4,7 +4,7 @@
\title{B-spline bases}
\description{
These functions generate B-spline basis functions.
- \code{bspline.basis} gives a set of basis functions.
+ \code{bspline.basis} gives a basis of spline functions.
\code{periodic.bspline.basis} gives a basis of periodic spline functions.
}
\usage{
@@ -27,7 +27,7 @@
The basis functions returned are periodic with period \code{period}.
}
}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\examples{
x <- seq(0,2,by=0.01)
y <- bspline.basis(x,degree=3,nbasis=9)
Modified: pkg/man/dmeasure-pomp.Rd
===================================================================
--- pkg/man/dmeasure-pomp.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/dmeasure-pomp.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -5,7 +5,9 @@
\alias{dmeasure-pomp}
\title{Evaluate the probability density of observations given underlying states in a partially-observed Markov process}
\description{
- The method \code{dmeaure} evaluates the probability density of a set of measurements given the state of the system.
+ The method \code{dmeasure} evaluates the probability density of a set of measurements given the state of the system.
+ This function is part of the low-level interface to \code{pomp} objects.
+ This help page does not give instructions on the implementation of models: see \code{\link{pomp}} for instructions.
}
\usage{
dmeasure(object, y, x, times, params, log = FALSE, \dots)
@@ -20,7 +22,6 @@
\item{x}{
a rank-3 array containing the states of the unobserved process.
The dimensions of \code{x} are \code{nvars} x \code{nreps} x \code{ntimes}, where \code{nvars} is the number of state variables, \code{nreps} is the number of replicates, and \code{ntimes} is the length of \code{times}.
- Note that if \code{ntimes != length(times)} or \code{ntimes != ncol(y)}, an error is generated.
}
\item{times}{
a numeric vector containing the times at which the observations were made.
@@ -29,7 +30,7 @@
a rank-2 array of parameters with columns corresponding to the columns of \code{x}.
Note that the \code{x} and \code{params} must agree in the number of their columns.
}
- \item{log}{if TRUE, probabilities p are given as log(p).}
+ \item{log}{if TRUE, log probabilities are returned.}
\item{\dots}{at present, these are ignored.}
}
\value{
@@ -40,8 +41,7 @@
This function is essentially a wrapper around the user-supplied \code{dmeasure} slot of the \code{pomp} object.
For specifications on writing such a function, see \code{\link{pomp}}.
}
-\references{}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{\code{\link{pomp-class}}, \code{\link{pomp}}}
\keyword{models}
\keyword{ts}
Modified: pkg/man/dprocess-pomp.Rd
===================================================================
--- pkg/man/dprocess-pomp.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/dprocess-pomp.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -8,6 +8,8 @@
}
\description{
The method \code{dprocess} evaluates the probability density of a set of consecutive state transitions.
+ This function is part of the low-level interface to \code{pomp} objects.
+ This help page does not give instructions on the implementation of models: see \code{\link{pomp}} for instructions.
}
\usage{
dprocess(object, x, times, params, log = FALSE, \dots)
@@ -18,7 +20,6 @@
\item{x}{
a rank-3 array containing the states of the unobserved process.
The dimensions of \code{x} are \code{nvars} x \code{nreps} x \code{ntimes}, where \code{nvars} is the number of state variables, \code{nreps} is the number of replicates, and \code{ntimes} is the length of \code{times}.
- Note that if \code{nreps != nrow(y)} or \code{ntimes-1 != length(times)}, an error is generated.
}
\item{times}{
a numeric vector containing the times corresponding to the given states.
@@ -27,7 +28,7 @@
a rank-2 array of parameters with columns corresponding to the columns of \code{x}.
Note that the \code{x} and \code{params} must agree in the number of their columns.
}
- \item{log}{if TRUE, probabilities p are given as log(p).}
+ \item{log}{if TRUE, log probabilities are returned.}
\item{\dots}{at present, these are ignored.}
}
\value{
@@ -38,7 +39,7 @@
This function is essentially a wrapper around the user-supplied \code{dprocess} slot of the \code{pomp} object.
For specifications on writing such a function, see \code{\link{pomp}}.
}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{\code{\link{pomp-class}}, \code{\link{pomp}}}
\keyword{models}
\keyword{ts}
Modified: pkg/man/euler.Rd
===================================================================
--- pkg/man/euler.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/euler.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -5,7 +5,8 @@
\alias{onestep.density}
\title{Plug-ins for dynamical models based on stochastic Euler algorithms}
\description{
- Facilities for implementing discrete-time Markov processes and continuous-time Markov processes using the Euler algorithm.
+ Plug-in facilities for implementing discrete-time Markov processes and continuous-time Markov processes using the Euler algorithm.
+ These can be used in the \code{rprocess} and \code{dprocess} slots of \code{pomp}.
}
\usage{
euler.simulate(xstart, times, params, step.fun, delta.t, \dots,
@@ -36,42 +37,33 @@
The \code{nrep} columns of \code{params} correspond to those of \code{xstart}.
}
\item{step.fun}{
- This can be either an R function or a compiled, dynamically loaded native function containing the model simulator.
+ This can be either an R function or the name of a compiled, dynamically loaded native function containing the model simulator.
It should be written to take a single Euler step from a single point in state space.
- If it is a native function, it must be of type "pomp\_onestep\_sim" as defined in the header "pomp.h", which is included with the package.
+ If it is a native function, it must be of type \dQuote{pomp\_onestep\_sim} as defined in the header \dQuote{pomp.h}, which is included with the package.
For details on how to write such codes, see Details.
}
\item{dens.fun}{
This can be either an R function or a compiled, dynamically loaded native function containing the model transition log probability density function.
This function will be called to compute the log likelihood of the actual Euler steps.
- It must be of type "pomp\_onestep\_pdf" as defined in the header "pomp.h", which is included with the package.
+ It must be of type \dQuote{pomp\_onestep\_pdf} as defined in the header \dQuote{pomp.h}, which is included with the package.
For details on how to write such codes, see Details.
}
\item{delta.t}{
Time interval of Euler steps.
}
- \item{statenames}{
- Names of state variables, in the order they will be expected by the routine named in \code{step.fun} and \code{dens.fun}.
+ \item{statenames, paramnames, covarnames}{
+ Names of state variables, parameters, covariates, in the order they will be expected by the routine named in \code{step.fun} and \code{dens.fun}.
+ This information is only used when the latter are implemented as compiled native functions.
}
- \item{paramnames}{
- Names of parameters, in the order they will be expected by the routine named in \code{step.fun} and \code{dens.fun}.
- }
- \item{covarnames}{
- Names of columns of the matrix of covariates \code{covar}, in the order they will be expected by the routine named in \code{step.fun} and \code{dens.fun}.
- }
\item{zeronames}{
Names of additional variables which will be zeroed before each time in \code{times}.
These are useful, e.g., for storing accumulations of state variables.
}
- \item{tcovar}{
- Times at which covariates are measured.
+ \item{covar, tcovar}{
+ Matrix of covariates and times at which covariates are measured.
}
- \item{covar}{
- Matrix of covariates.
- This should have dimensions \code{length(tcovar)} x \code{ncovar}, where \code{ncovar} is the number of covariates.
- }
\item{log}{
- logical; if TRUE, probabilities p are given as log(p).
+ logical; if TRUE, log probabilities are given.
}
\item{\dots}{
if \code{step.fun} (or \code{dens.fun}) is an R function, then additional arguments will be passed to it.
@@ -113,7 +105,7 @@
\code{onestep.density} returns a \code{nrep} x \code{ntimes-1} array.
If \code{f} is this array, \code{f[i,j]} is the likelihood of a transition from \code{x[,i,j]} to \code{x[,i,j+1]} in exactly one Euler step of duration \code{times[j+1]-times[j]}.
}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{\code{\link{eulermultinom}}, \code{\link{pomp}}}
\examples{
## an example showing how to use these functions to implement a seasonal SIR model is contained
Modified: pkg/man/euler.sir.Rd
===================================================================
--- pkg/man/euler.sir.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/euler.sir.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -6,8 +6,6 @@
\code{euler.sir} is a \code{pomp} object encoding a simple seasonal SIR model.
}
\usage{data(euler.sir)}
-\details{
-}
\examples{
data(euler.sir)
plot(euler.sir)
Modified: pkg/man/eulermultinom.Rd
===================================================================
--- pkg/man/eulermultinom.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/eulermultinom.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -4,20 +4,23 @@
\alias{deulermultinom}
\title{Euler-multinomial models}
\description{
- Density and random generation for the Euler-multinomial death process with parameters \code{size}, \code{rate}, and \code{dt}.
+ Density and random-deviate generation for the Euler-multinomial death process with parameters \code{size}, \code{rate}, and \code{dt}.
}
\usage{
reulermultinom(n = 1, size, rate, dt)
deulermultinom(x, size, rate, dt, log = FALSE)
}
\arguments{
- \item{n}{Number of random variates to generate.}
- \item{size}{Number of individuals at risk.}
- \item{rate}{Hazard rates.}
- \item{dt}{Duration of Euler step.}
+ \item{n}{integer; number of random variates to generate.}
+ \item{size}{scalar integer; number of individuals at risk.}
+ \item{rate}{numeric vector of hazard rates.}
+ \item{dt}{numeric scalar; duration of Euler step.}
\item{x}{Matrix or vector containing number of individuals that have succumbed to each death process.}
- \item{log}{logical; if TRUE, return logarithm of probabilities.}
+ \item{log}{logical; if TRUE, return logarithm(s) of probabilities.}
}
+\details{
+ Direct access to the underlying C routines is available: see the header file \dQuote{pomp.h}, included with the package.
+}
\value{
\item{reulermultinom}{
Returns a \code{length(rate)} by \code{n} matrix.
@@ -25,15 +28,11 @@
Each row contains the numbers of individuals succumbed to the corresponding process.
}
\item{deulermultinom}{
- Returns a vector (of length equal to the number of columns of \code{x}) containing the probabilities of observing \code{x} given the specified parameters (\code{size}, \code{rate}, \code{dt}).
+ Returns a vector (of length equal to the number of columns of \code{x}) containing the probabilities of observing each column of \code{x} given the specified parameters (\code{size}, \code{rate}, \code{dt}).
}
}
-\details{
- Direct access to the underlying C routines is available: see the header file "pomp.h", included with the package,
- e.g., \code{edit(file=system.file("include/pomp.h",package="pomp"))}.
-}
-\author{Aaron A. King (kingaa at umich dot edu)}
-\seealso{\code{\link{euler}}, \code{\link{pomp}}}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
+\seealso{\code{\link{euler}}}
\examples{
print(x <- reulermultinom(5,size=100,rate=c(a=1,b=2,c=3),dt=0.1))
deulermultinom(x,size=100,rate=c(1,2,3),dt=0.1)
Modified: pkg/man/init.state-pomp.Rd
===================================================================
--- pkg/man/init.state-pomp.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/init.state-pomp.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -3,12 +3,11 @@
\alias{init.state}
\alias{init.state,pomp-method}
\alias{init.state-pomp}
-
-\title{
- Return a matrix of initial conditions given a vector of parameters and an initial time.
-}
+\title{Return a matrix of initial conditions given a vector of parameters and an initial time.}
\description{
The method \code{init.state} returns a vector of initial conditions for the state process when given a vector of parameters \code{params} and an initial time \code{t0}.
+ This function is part of the low-level interface to \code{pomp} objects.
+ This help page does not give instructions on the implementation of models: see \code{\link{pomp}} for instructions.
}
\usage{
init.state(object, params, t0, \dots)
@@ -25,8 +24,7 @@
\value{
Returns a matrix of initial states (with rownames).
}
-\references{}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{\code{\link{pomp-class}}}
\keyword{models}
\keyword{ts}
Modified: pkg/man/mif-class.Rd
===================================================================
--- pkg/man/mif-class.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/mif-class.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -3,11 +3,11 @@
\alias{mif-class}
\title{The "mif" class}
\description{
- The MIF algorithm: maximum likelihood via iterated filtering.
- The \code{mif} class holds a fitted model.
+ The \code{mif} class holds a fitted model and is created by a call to \code{\link{mif}}.
+ See \code{\link{mif}} for usage.
}
\section{Objects from the Class}{
- Objects can be created by calls to the \code{mif} method on an \code{pomp} object.
+ Objects can be created by calls to the \code{\link{mif}} method on an \code{\link{pomp}} object.
Such a call uses the MIF algorithm to fit the model parameters.
}
\section{Slots}{
@@ -28,7 +28,6 @@
\item{particles}{
A function of prototype \code{particles(Np,center,sd,...)} that draws particles from a distribution centered on \code{center} and with width proportional to \code{sd}.
This function can be optionally specified by the user.
- Its default value is a multivariate normal distribution with mean at \code{center} and standard deviation \code{sd}.
}
\item{alg.pars}{
A named list of algorithm parameters.
@@ -36,11 +35,10 @@
\code{Np}, the number of particles to use in filtering;
\code{var.factor}, the scaling coefficient relating the width of the initial particle distribution to \code{rw.sd};
\code{ic.lag}, the fixed lag used in the estimation of initial-value parameters (IVPs);
- and \code{cooling.factor}, the exponential cooling factor, \code{alpha}, where \code{0<alpha<1}.
+ and \code{cooling.factor}, the exponential cooling factor, where \code{0<cooling.factor<1}.
}
\item{random.walk.sd}{
A named vector containing the random-walk variance to be used for ordinary parameters.
- The width of the initial distribution of particles will be random.walk.sd*var.factor.
}
\item{pred.mean}{
Matrix of prediction means.
@@ -63,7 +61,7 @@
See \code{\link{pfilter}}.
}
\item{conv.rec}{
- The "convergence record": a matrix containing a record of the parameter values, log likelihoods, and other pertinent information, with one row for each MIF iteration.
+ The \dQuote{convergence record}: a matrix containing a record of the parameter values, log likelihoods, and other pertinent information, with one row for each MIF iteration.
}
\item{loglik}{
A numeric value containing the value of the log likelihood, as evaluated for the random-parameter model.
@@ -93,7 +91,7 @@
Inapparent infections and cholera dynamics,
Nature, 454:877--880, 2008.
}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{\code{\link{mif}}, \link{mif-methods}, \code{\link{pomp}}, \link{pomp-class}}
\keyword{models}
\keyword{ts}
Modified: pkg/man/mif-methods.Rd
===================================================================
--- pkg/man/mif-methods.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/mif-methods.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -54,7 +54,7 @@
}
\item{mif}{
Re-runs the MIF iterations.
- See the documentation under \cite{Re-running MIF Iterations} below.
+ See the documentation for \code{\link{mif}}.
}
\item{compare.mif}{
Given a \code{mif} object or a list of \code{mif} objects, \code{compare.mif} produces a set of diagnostic plots.
@@ -101,7 +101,7 @@
Inapparent infections and cholera dynamics,
Nature, 454:877--880, 2008.
}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{\code{\link{mif}}, \code{\link{pomp}}, \code{\link{pomp-class}}, \code{\link{pfilter}}}
\keyword{models}
\keyword{ts}
Modified: pkg/man/mif.Rd
===================================================================
--- pkg/man/mif.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/mif.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -72,6 +72,7 @@
\item{var.factor}{
a positive number;
the scaling coefficient relating the width of the initial particle distribution to \code{rw.sd}
+ The width of the initial distribution of particles will be \code{random.walk.sd*var.factor}.
}
\item{cooling.factor}{
a positive number not greater than 1;
@@ -103,7 +104,7 @@
\section{Re-running MIF Iterations}{
To re-run a sequence of MIF iterations, one can use the \code{mif} method on a \code{mif} object.
The call sequence is \code{mif(object)}.
- By default, the same parameters used for the original MIF run are re-used.
+ By default, the same parameters used for the original MIF run are re-used (except for \code{weighted}, \code{tol}, \code{warn}, \code{max.fail}, and \code{verbose}, the defaults of which are shown above).
If one does specify additional arguments, these will override the defaults.
}
\section{Continuing MIF Iterations}{
@@ -115,17 +116,18 @@
Additional arguments will override the defaults.
}
\section{Details}{
+ If \code{particles} is not specified, the default behavior is to draw the particles from a multivariate normal distribution.
\strong{It is the user's responsibility to ensure that, if the optional \code{particles} argument is given, that the \code{particles} function satisfies the following conditions:}
\code{particles} has at least the following arguments:
\code{Np}, \code{center}, \code{sd}, and \code{\dots}.
- \code{Np} should be assumed to be an integer; \code{center} and \code{sd} will be named vectors of the same length.
+ \code{Np} may be assumed to be a positive integer;
+ \code{center} and \code{sd} will be named vectors of the same length.
Additional arguments may be specified;
these will be filled with the elements of the \code{userdata} slot of the underlying \code{pomp} object (see \code{\link{pomp-class}}).
- \code{particles} returns a \code{length(center)} x \code{Np} matrix with rownames.
+ \code{particles} returns a \code{length(center)} x \code{Np} matrix with rownames matching the names of \code{center} and \code{sd}.
Each column represents a distinct particle.
- The rownames are used by the algorithms (see \code{mif}, \code{pfilter}).
The center of the particle distribution returned by \code{particles} should be \code{center}.
The width of the particle distribution should vary monotonically with \code{sd}.
@@ -140,10 +142,10 @@
Inapparent infections and cholera dynamics,
Nature, 454:877--880, 2008.
}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{
\code{\link{mif-class}}, \code{\link{mif-methods}}, \code{\link{pomp}}, \code{\link{pomp-class}}, \code{\link{pfilter}}.
- See the "intro\_to\_pomp" vignette for an example.
+ See the \dQuote{intro\_to\_pomp} vignette for an example.
}
\keyword{models}
\keyword{ts}
Modified: pkg/man/nlf.Rd
===================================================================
--- pkg/man/nlf.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/nlf.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -3,12 +3,12 @@
\title{Fit Model to Data Using Nonlinear Forecasting (NLF)}
\description{
Calls an optimizer to maximize the nonlinear forecasting (NLF) goodness of fit, by simulating data from a model, fitting a nonlinear autoregressive model to the simulated time series (which may be multivariate) and using the fitted model to predict some or all variables in the data time series.
- NLF is an 'indirect inference' method using a quasi-likelihood as the objective function.
+ NLF is an \sQuote{indirect inference} method using a quasi-likelihood as the objective function.
}
\usage{
nlf(object, start, est, lags, period = NA, tensor = FALSE,
nconverge=1000, nasymp=1000, seed = 1066, nrbf = 4,
- method = 'subplex', skip.se = FALSE, verbose = FALSE, gr = NULL,
+ method = "subplex", skip.se = FALSE, verbose = FALSE, gr = NULL,
bootstrap=FALSE, bootsamp = NULL,
lql.frac = 0.1, se.par.frac = 0.1, eval.only = FALSE, \dots)
}
@@ -76,14 +76,14 @@
logical; if \code{TRUE}, no optimization is attempted and the quasi-loglikelihood value is evaluated at the \code{start} parameters.
}
\item{\dots}{
- Arguments that will be passed to \code{optim} in the \code{control} list.
+ Arguments that will be passed to \code{optim} or \code{subplex} in the \code{control} list.
}
}
\details{
This is functionally a wrapper for \code{nlf.objfun}, which does the statistical heavy lifting and should be consulted for details.
}
\value{
- A list corresponding to the output from the optimizer, except that the full parameter vector is returned (not just the ones fitted), the LQL (and not -LQL) is reported, xstart is included, and asymptotic Wald standard errors based on M-estimator theory are returned for each fitted parameter.
+ A list corresponding to the output from the optimizer, except that the full parameter vector is returned (not just the ones fitted), the log quasilikelihood (LQL) (\emph{not} -LQL) is reported, xstart is included, and asymptotic Wald standard errors based on M-estimator theory are returned for each fitted parameter.
}
\references{
The following papers describe and motivate the NLF approach to model fitting:
@@ -102,7 +102,7 @@
\emph{Ecological Monographs} \bold{75}, 259--276.
Available online at \url{http://repositories.cdlib.org/postprints/818/}
}
-\author{Stephen P. Ellner (spe2 at cornell dot edu) and Bruce E. Kendall (kendall at bren dot ucsb dot edu)}
+\author{Stephen P. Ellner \email{spe2 at cornell dot edu} and Bruce E. Kendall \email{kendall at bren dot ucsb dot edu}}
\keyword{models}
\keyword{ts}
Modified: pkg/man/ou2.Rd
===================================================================
--- pkg/man/ou2.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/ou2.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -3,7 +3,7 @@
\docType{data}
\title{Two-dimensional Ornstein-Uhlenbeck process}
\description{
- \code{ou2} is a \code{pomp} object encoding a 2-D Ornstein-Uhlenbeck process.
+ \code{ou2} is a \code{pomp} object encoding a bivariate Ornstein-Uhlenbeck process.
}
\usage{data(ou2)}
\details{
@@ -15,6 +15,7 @@
\examples{
data(ou2)
plot(ou2)
+coef(ou2)
p <- c(
alpha.1=0.9,alpha.2=0,alpha.3=0,alpha.4=0.99,
sigma.1=1,sigma.2=0,sigma.3=2,
Modified: pkg/man/particles-mif.Rd
===================================================================
--- pkg/man/particles-mif.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/particles-mif.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -4,13 +4,18 @@
\alias{particles,mif-method}
\alias{particles-mif}
\title{Generate particles from the user-specified distribution.}
-\description{Generate particles from the user-specified distribution.}
+\description{
+ Generate particles from the user-specified distribution.
+ This is part of the low-level interface, used by \code{\link{mif}}.
+ This help page does not give instruction on how to write a valid \code{particles} function:
+ see the documentation for \code{\link{mif}} instead.
+}
\usage{
particles(object, \dots)
\S4method{particles}{mif}(object, Np = 1, center = coef(object), sd = 0, \dots)
}
\arguments{
- \item{object}{the "mif" object}
+ \item{object}{the \code{mif} object}
\item{Np}{the number of particles, i.e., number of draws.}
\item{center}{the central value of the distribution of particles}
\item{sd}{the width of the distribution}
@@ -19,15 +24,14 @@
\details{
The \code{particles} method is used to set up the initial distribution
of particles. It is an interface to the user-specifed
- \code{particles} slot in the "mif" object.
+ \code{particles} slot in the \code{mif} object.
}
\value{
\code{particles} returns a list of two matrices. \code{states}
contains the state-variable portion of the particles; \code{params}
contains the parameter portion. Each has \code{Np} columns.
}
-\author{Aaron A. King (kingaa at umich dot edu)}
-\seealso{\code{\link{mif}}, \link{mif-methods}, \code{\link{pomp}},
- \link{pomp-class}}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
+\seealso{\code{\link{mif}}, \link{mif-methods}, \code{\link{pomp}}, \link{pomp-class}}
\keyword{models}
\keyword{ts}
Modified: pkg/man/pfilter.Rd
===================================================================
--- pkg/man/pfilter.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/pfilter.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -6,7 +6,8 @@
\alias{pfilter-mif}
\title{Particle filter}
\description{
- Run a particle filter.
+ Run a plain vanilla particle filter.
+ Resampling is performed after each observation.
}
\usage{
pfilter(object, \dots)
@@ -22,7 +23,7 @@
An object of class \code{pomp} or inheriting class \code{pomp}.
}
\item{params}{
- A \code{npars} x \code{np} matrix containing the parameters corresponding to the initial state values in \code{xstart}.
+ A \code{npars} x \code{Np} matrix containing the parameters corresponding to the initial state values in \code{xstart}.
This must have a 'rownames' attribute.
It is permissible to supply \code{params} as a named numeric vector, i.e., without a \code{dim} attribute.
In this case, all particles will inherit the same parameter values.
@@ -92,7 +93,7 @@
A Tutorial on Particle Filters for Online Nonlinear, Non-Gaussian Bayesian Tracking.
IEEE Trans. Sig. Proc. 50:174--188, 2002.
}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{\link{pomp-class}}
\keyword{models}
\keyword{ts}
Modified: pkg/man/pomp-class.Rd
===================================================================
--- pkg/man/pomp-class.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/pomp-class.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -2,11 +2,13 @@
\docType{class}
\alias{pomp-class}
\title{Partially-observed Markov process}
-\description{The class \code{pomp} encodes a partially-observed Markov process.}
+\description{
+ The class \code{pomp} encodes a partially-observed Markov process.
+ This page documents the structure of the class:
+ see the documentation for \code{\link{pomp}} for usage instructions.
+}
\section{Objects from the Class}{
- Objects should be created by calls of the function
- \code{pomp}. See the documentation for \code{\link{pomp}} for
- usage instructions and important warnings.
+ Objects should be created by calls of the function \code{pomp}.
}
\section{Slots}{
\describe{
@@ -26,16 +28,16 @@
Function of prototype \code{dprocess(x,times,params,log=FALSE,\dots)} which evaluates the likelihood of a sequence of consecutive state transitions.
}
\item{dmeasure}{
- an object of class "pomp.fun" which encodes the measurement model density.
+ an object of class \dQuote{pomp.fun} which encodes the measurement model density.
}
\item{rmeasure}{
- an object of class "pomp.fun" which encodes the measurement model simulator.
+ an object of class \dQuote{pomp.fun} which encodes the measurement model simulator.
}
\item{skeleton.type}{
a character variable specifying whether the deterministic skeleton is a map or a vectorfield.
}
\item{skeleton}{
- an object of class "pomp.fun" which encodes the deterministic skeleton.
+ an object of class \dQuote{pomp.fun} which encodes the deterministic skeleton.
}
\item{initializer}{
Function of prototype \code{initializer(params,t0,\dots)} which gives a vector of initial conditions when given a vector of parameters, \code{params}, and a time \code{t0}.
@@ -65,7 +67,7 @@
\section{Methods}{
See the pomp methods documentation: \link{pomp-methods}.
}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{
\code{\link{pomp}},
\link{pomp-methods},
Modified: pkg/man/pomp-methods.Rd
===================================================================
--- pkg/man/pomp-methods.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/pomp-methods.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -62,7 +62,7 @@
name of the class to which \code{object} should be coerced.
}
\item{from, to}{
- the classes betwen which coercion should be performed.
+ the classes between which coercion should be performed.
}
\item{strict}{
ignored.
@@ -73,7 +73,7 @@
names of variables to plot.
}
\item{panel}{
- a 'function(x, col, bg, pch, type, ...)' which gives the action to be carried out in each panel of the display.
+ a function of prototype \code{panel(x, col, bg, pch, type, ...)} which gives the action to be carried out in each panel of the display.
}
\item{nc}{
the number of columns to use.
@@ -114,7 +114,7 @@
}
\item{data.array}{
\code{data.array(object)} returns the array of observations.
- \code{data.array(object,vars)} gives just the observations of variables \code{vars}.
+ \code{data.array(object,vars)} gives just the observations of variables named \code{vars}.
\code{vars} may specify the variables by position or by name.
}
\item{states}{
@@ -168,7 +168,7 @@
}
}
}
-\author{Aaron A. King (kingaa at umich dot edu)}
+\author{Aaron A. King \email{kingaa at umich dot edu}}
\seealso{
\code{\link{pomp}},
\link{pomp-class},
Modified: pkg/man/pomp-package.Rd
===================================================================
--- pkg/man/pomp-package.Rd 2009-06-01 17:30:52 UTC (rev 137)
+++ pkg/man/pomp-package.Rd 2009-06-02 19:20:25 UTC (rev 138)
@@ -3,31 +3,29 @@
\alias{pomp-package}
\title{Partially-observed Markov processes}
\description{
- The \code{pomp} package provides facilities for inference using partially-observed Markov processes (AKA state-space models or nonlinear stochastic dynamical systems).
- The user provides functions specifying some or all of the model's process and measurement components.
- The package's algorithms are built on top of these functions.
+ The \code{pomp} package provides facilities for inference on time series data using partially-observed Markov processes (AKA state-space models or nonlinear stochastic dynamical systems).
+ The user encodes a model as a \code{\link{pomp}} object by providing functions specifying some or all of the model's process and measurement components.
+ The package's algorithms for fitting models to data, simulating, etc. then call these functions.
At the moment, algorithms are provided for
- particle filtering (AKA sequential Monte Carlo or sequential importance sampling),
- the likelihood maximization by iterated filtering (MIF) method of Ionides, Breto, and King (PNAS, 103:18438-18443, 2006),
- and the nonlinear forecasting algorithm of Kendall, Ellner, et al. (Ecol. Monog. 75:259-276, 2005).
+ particle filtering (AKA sequential Monte Carlo or sequential importance sampling, see \code{\link{pfilter}}),
+ the likelihood maximization by iterated filtering (\acronym{MIF}) method of Ionides, Breto, and King (PNAS, 103:18438-18443, 2006, see \code{\link{mif}}),
+ and the nonlinear forecasting algorithm of Kendall, Ellner, et al. (Ecol. Monog. 75:259-276, 2005, see \code{\link{nlf}}).
Future support for a variety of other algorithms is envisioned.
- A working group of the National Center for Ecological Analysis and Synthesis (NCEAS), "Inference for Mechanistic Models", is currently implementing additional methods for this package.
[TRUNCATED]
To get the complete diff run:
svnlook diff /svnroot/pomp -r 138
More information about the pomp-commits
mailing list