NEWS | R Documentation |
News for package ‘pomp’
Changes in pomp version 5.10.1
An improper typecast in the C snippet system, which was resolvable by many compilers but threw an error under recent versions of clang on MacOS, has been repaired.
Changes in pomp version 5.9.1
The rprocess plugin
onestep
now always takes exactly one step to get from one observation time to the next, even when the interval between them is zero. That is, ifP
is a ‘pomp’ object with an rprocess built usingrprocess=onestep(f)
,t0=timezero(P)
, andt=time(P)
, thenf
will be called exactly once for the interval (t0
,t[1]
) and once for each of the intervals (t[i-1]
,t[i]
),i=2,...,length(t)
.
Changes in pomp version 5.8.4
To add additional elements for use by the basic model components (i.e., “userdata”), the
userdata
argument should now be used. The old behavior will still work, but will generate a warning. In a future release, this will become an error.
Changes in pomp version 5.8.3
Internally, pomp codes no longer include ‘Rinternal.h’.
Changes in pomp version 5.8.2
A bug in covariate-table extrapolation for the case
order="constant"
has been fixed.A new experimental function
repair_lookup_table
is provided to help eliminate unnecessary warnings about extrapolation.
Changes in pomp version 5.8.1
Functions with names that include dots (
.
), which have been defunct since version 4.7, have been expunged.
Changes in pomp version 5.7.1
The
err()
andwarn()
C-level macros have been redefined to forestall problems that would arise when these macros are included in C++ code under forthcoming versions of R.More changes in C codes to forestall issues detected by rchk.
Minor corrections to documentation.
Changes in pomp version 5.6.1
Minor corrections to documentation.
Changes in some C codes to avoid issues detected by rchk.
Fixed bug in
wquant
introduced by changes inpbeta
with R version 4.3.3.
Changes in pomp version 5.5.2
Minor corrections to documentation.
Changes in pomp version 5.5.1
C-level functions
set_pomp_userdata
andunset_pomp_userdata
are no longer exported.-
bspline_eval
is no longer exported.
Changes in pomp version 5.4.4
C-level functions
set_pomp_userdata
andunset_pomp_userdata
have been temporarily restored. The latter generates a warning.
Changes in pomp version 5.4.3
There is now access to ordinary B-spline basis functions (and their derivatives) at the C snippet level. It will still almost always be preferable to construct a spline basis once and pass it to pomp functions as part of a
covariate_table
, but this functionality may occasionally be useful.-
bspline_eval
is now deprecated as part of the pomp C API. The formatting of some printed messages has been modified to forestall warnings in the latest R-devel.
Changes in pomp version 5.4.2
Stateful objective functions (class ‘objfun’) now behave as if they are ‘pomp’ objects.
Changes in pomp version 5.4.1
C-level functions
set_pomp_userdata
andunset_pomp_userdata
are no longer exported.
Changes in pomp version 5.3.1
-
stew
now returns timing information by default. The spurious warnings associated with NULL
userdata
have been eliminated.
Changes in pomp version 5.2.5
The new option
on_load
allows one to specify a C snippet that will be executed at the time the C snippet library is loaded.Internally, the
userdata
are made available via calls topompLoad
rather than within each call to a pomp workhorse.
Changes in pomp version 5.2.4
The data frames returned by
cond_logLik
andeff_sample_size
whenformat="data.frame"
have been improved. In particular, they contain (as variabletime
) the times (rather than the index of the time vector as before).Changes to the report generated by
spy
.Some documentation improvements.
Changes in pomp version 5.2.3
The manual pages have been reorganized, with improved cross-linking.
A bug in
filter_traj
etc. arising when not all state variables have names was fixed.
Changes in pomp version 5.2.2
This version adds the new basic component
dinit
for evaluating the probability density function of the initial-state distribution. There is a corresponding workhorse function,dinit()
.The constructor
pomp
now takes the optional argumentnstatevars
, which can be used to increase the dimension of the latent state-vectors created byrinit
. By default,nstatevars = length(statenames)
, andnstatevars
can only be used to increase, not to decrease, the dimension of the latent state process. Moreover,nstatevars
has no effect if therinit
basic component is furnished as an R function.Name checking on internally-created vectors and arrays is less strict than previously. In particular, it is possible to have variables without names (i.e.,
""
).
Changes in pomp version 5.2.1
-
reulermultinom
now returnsNA
rather thanNaN
, in keeping with the behavior ofrbinom
. Thanks to John Drake for calling attention to this issue.
Changes in pomp version 5.1.1
A bug in
stew
resulting in namespace conflicts has been repaired.We no longer import directly from tidyverse functions. This reduces the list of packages upon which pomp depends from 20 to 5. This does introduce a dependency on the data.table package.
Changes in pomp version 5.0.1
When the archive directory does not exist,
bake
andstew
now create it.
Changes in pomp version 5.0.0
We no longer import from plyr.
The
dimnames
attributes of various arrays that appear in pomp, including arrays of observables, state variables, parameters, covariates, and so on, have been made uniform. In particular, when a dimension of an array corresponds to variables with different names, this dimension is itself named “name”. Previously, its name was sometimes “variable” and sometimes “parameter”. This change is meant to streamline interaction with the tidyverse.Some minor bugfixes.
Changes in pomp version 4.7.0
Functions with names.that.contain.dots that have been deprecated since version 4.5.1 are now defunct. Attempts to use such functions will generate an error with a message indicating the replacement function.
-
plot
now removesNA
.
Changes in pomp version 4.6.4
Better handling of ‘listie’s (lists of ‘pomp’ objects).
New
concat
function to turn lists of ‘pomp’ objects into ‘listies’.
Changes in pomp version 4.6.3
Bug fix in
filter_traj
applied to ‘pfilterList’ and ‘pmcmcList’ objects.
Changes in pomp version 4.6.2
The
states
andobs
methods take a new optional argument,format
. Settingformat="data.frame"
causes the method to return the states or data in a convenient data-frame format.
Changes in pomp version 4.6.1
When
melt
is applied to a list, the identifier variable is now “.Lx
”, wherex
denotes the level. This applies whenas.data.frame
is used to coerce a list of ‘pomp’ objects, and whensaved_states
is used to extract the particles from a list of ‘pfilterd_pomp’ objects.
Changes in pomp version 4.5.3
-
logmeanexp
can now compute the effective sample size, which can be useful in determining the reliability of its estimate. To do so, setess=TRUE
in a call tologmeanexp
. -
logmeanexp
now returns a fully-named vector when eitherse=TRUE
oress=TRUE
.
Changes in pomp version 4.5.2
A new
format
argument has been added to the extractor functionscond_logLik
,eff_sample_size
,filter_mean
,filter_traj
,forecast
,pred_mean
,pred_var
, andsaved_states
. This allows the user to extract the relevant elements in data-frame format if desired.-
pomp no longer depends on the superseded package reshape2. The
melt
function for converting arrays and nested lists into data frames is accordingly no longer re-exported from reshape2. It has been replaced by a stricter version ofmelt
. The magrittr pipe
%>%
is no longer re-exported by pomp: use the native R pipe|>
instead.The package now requires R version 4.1 at least.
Changes in pomp version 4.5.1
All pomp functions with names.that.contain.dots have been deprecated in favor of functions in ‘snake_case’. This is to avoid anticipated problems with CRAN checks, which (falsely) assume that certain functions with dotted.names are S3 methods. From this point, no exported pomp function has such a name.
The
dimnames
attributes for some of the arrays computed inpfilter
andpmcmc
computations have changed. In particular, whereas in previous versions, thetime
dimension was given names that were character strings composed of decimal representations of the time (difficult to work with and prone to roundoff error), thetime
dimension now is not given names.A bug that resulted in
gompertz
giving different simulations on Windows machines has been fixed.
Changes in pomp version 4.4.3
A bug in trajectory computation for maps (discrete-time dynamical systems) has been fixed. Thanks to Felicia Magpantay for noticing it and tracking it down.
Minor documentation improvements.
Changes in pomp version 4.4.2
The weighted quantile function
wquant
now uses the Harrell-Davis estimator instead of the type-7 estimator used previously. This eliminates bugs associated with non-integer weights.
Changes in pomp version 4.4.1
It is now possible to retrieve the weighted particles computed in the course of a
pfilter
computation. To accomplish this, setsave.states="weighted"
in the call topfilter
and retrieve the particles and their weights usingsaved.states
. Previously, one could obtain only the unweighted particles.
Changes in pomp version 4.4.0
Some documentation improvements.
Changes in pomp version 4.3.4
The new function
wquant
computes weighted quantiles.
Changes in pomp version 4.3.3
The precise order in which the
pmcmc
function computes prior and likelihood of furnished and proposed parameters has changed slightly. This prevents proposals that are incompatible with the prior from being passed topfilter
and forestalls an associated class of errors. A consequence of this change,pmcmc
computations using the new version will differ very slightly from previous computations, even with the RNG seed fixed at its previous values.
Changes in pomp version 4.3.2
There is now a C interface to the
bspline_eval
function. See the pomp C API documentation for details.
Changes in pomp version 4.3.1
The
bspline.basis
function now takes the optional argumentrg
which allows one to specify the range over which the basis will be constructed. By default, this isrange(x)
, which agrees with the behavior in earlier versions.Documentation improvements.
Changes in pomp version 4.2.2
The archiving functions
bake
andstew
now use a slightly less exacting comparison of the expression,expr
, they are furnished. NB: Runningbake
orstew
with archives created by earlier versions may result in recomputation.The
time
method has been extended topompList
and related objects.All workhorse functions except
partrans
have new default arguments.Some documentation improvements.
Changes in pomp version 4.2.1
The archiving functions
bake
andstew
now take the argumentdir
, which is the directory holding the archive files. By default, this is the current working directory or the value of the global optionpomp.archive.dir
.A new global option,
pomp_archive_dir
, can be used to specify the locations of archive files used bybake
andstew
.Documentation repair of issue flagged by Kurt Hornik (9 June 2022).
Changes in pomp version 4.2.0
Minor changes, invisible to the user.
Minor changes to the C codes to keep up with R-devel.
Changes in pomp version 4.1.0
Minor changes to the C codes to keep up with R-devel.
Minor changes in the implementation of
bake
andstew
.
Changes in pomp version 4.0.11
Change in the names of the variables returned by
as.data.frame
applied to a ‘pfilterd.pomp’ object.Fixed bug found by Jesse Wheeler and Ed Ionides, to do with 'simulate()' with 'include.data=TRUE'.
Changes in pomp version 4.0.10
It is now possible to plot lists of ‘pomp’ and ‘pomp’-derived objects using a single call to
plot
.
Changes in pomp version 4.0.9
The
rbetabinom
anddbetabinom
functions, long present as part of the C API, are now available as R functions.
Changes in pomp version 4.0.8
The SIR examples
sir()
andsir2()
now use a negative binomial measurement model.
Changes in pomp version 4.0.7
-
parmat
can now take a data frame of parameters and convert it into a matrix suitable for furnishing to theparams
argument of any pomp function.
Changes in pomp version 4.0.6
It is now possible to change the parameters in an ‘objfun’ (objective function) object using
coef(object)<-value
.
Changes in pomp version 4.0.5
-
parmat
now takes an optional argument,names
, which allows the user to name the parameter sets. Independent realizations of the stochastic processes modeled in pomp are now distinguished by different values of
.id
. This behavior is now uniform throughout.-
bake
andstew
now send messages instead of warnings when they recompute an archive due to a change in code or dependencies.
Changes in pomp version 4.0.4
The ensemble adjusted Kalman filter (
eakf
) has been refactored. It now makes use of theemeasure
andvmeasure
basic components to compute an approximation of the linear relationship between the latent state and the observed variables.The
forecast
method now works forpfiterd_pomp
objects (i.e., results ofpfilter
,pmcmc
, ormif2
computations).
Changes in pomp version 4.0.3
Two new basic model components have been introduced. These are:
- emeasure,
which computes the expectation of the observable variables conditional on the latent state;
- vmeasure,
which computes the covariance matrix of the observables given the latent state.
These are used by the newly refactored ensemble Kalman filter,
enkf
.The basic (linear, Gaussian) Kalman filter is now available as
kalmanFilter
.
Changes in pomp version 4.0.2
Computation of Monte Carlo-adjusted profile likelihood is now facilitated by the new
mcap
function.
Changes in pomp version 4.0.1
The default return-value format for
trajectory
now matches that ofsimulate
.
Changes in pomp version 4.0.0
Version 4 contains a few changes that are not backward compatible.
These are all to do with the elementary function trajectory
, which computes trajectories of the deterministic dynamical skeleton.
See below for details.
The behavior of
trajectory
now fully conforms to the behavior of other pomp elementary functions. In particular, one can now add, remove, or modify basic model components in a call totrajectory
just as one can withsimulate
,pfilter
,probe
, etc. Before version 4, additional arguments totrajectory
(i.e., those passed via...
) were passed on to the ODE integrator in the case of continuous-time deterministic skeletons (i.e., vectorfields) and ignored in the case of discrete-time skeletons (i.e., maps). As of version 4, in order to adjust ODE integrator settings it is necessary to use theode_control
argument oftrajectory
. This behavior matches that oftraj_objfun
.It is now possible to create a ‘pomp’ object from scratch using
trajectory
, together with a specification of the rinit and skeleton components. Prior to version 4, in order to do so, it was necessary to first create a dummy data set, then callpomp
, and then pass the resulting ‘pomp’ object totrajectory
. This can now be achieved in one call.The user now has the option, in a call to
trajectory
, to have the results returned as one or ‘pomp’ objects. In this regard, its behavior matches that ofsimulate
.
Changes in pomp version 3.5.6
The
states
andobs
methods now work for lists of ‘pomp’ objects (‘listies’).
Changes in pomp version 3.5.5
The workhorse function
flow
now has default arguments.The stateful objective function created by
traj_objfun
has been refactored so that it is independent oftrajectory
. This change is invisible to the user, but lays the groundwork for future changes intrajectory
.
Changes in pomp version 3.5.4
Stateful objective functions created by
traj_objfun
,nlf_objfun
,probe_objfun
, andspect_objfun
now have default arguments. In particular, the default is argument is a zero-length numeric vector. Calling such a function with no arguments is valid if and only if the objective function was created with no variables to estimate (emptyest
).
Changes in pomp version 3.5.3
Fixed bug in
rprior
arising with integer-valued return vectors.
Changes in pomp version 3.5.2
Major improvements to the package manual, including richer cross-linking and better organization.
Changes in pomp version 3.5.1
When
simulate
is called withformat="data.frame"
andinclude.data=TRUE
, the interpolated covariates are now included in the data frame that is returned.
Changes in pomp version 3.4.6
-
bake
takes a new option,timing
, that controls whether information on the timing is returned or merely stored. By default, this is set toTRUE
, though this default behavior may change in future.
Changes in pomp version 3.4.5
When they encounter an archive produced by an older version,
bake
andstew
now update the archive in the new format. Recomputation should never be triggered in this case: If recomputation is desired, delete the archive file.
Changes in pomp version 3.4.4
-
bake
andstew
now archive a digest of the code inexpr
and on subsequent calls check to see if the archived version matches the current version. Thus, changes to the code automatically trigger recomputation. -
bake
andstew
take the new argumentdependson
. Using this argument, the user can specify objects upon which the computation depends. These should be passed as unquoted variables; if there are multiple dependencies, they can be passed usingc
orlist
. Digests of these objects will be archived and checked against current values on subsequent calls. Thus, changing any dependency will automatically trigger recomputation. -
bake
now archives the information pertinent to thebake
call; this information is no longer returned along with the result of the computation (unlessinfo=TRUE
). Similarly,stew
stores a list, “.ingredients”, in the file it creates. In thestew
case, unlike all the other objects that are created byexpr
, this list is not by default loaded into the calling environment (e.g., the user's workspace). It will be loaded if theinfo
flag is set toTRUE
. -
freeze
no longer stores the details of the RNG seed and kind as an attribute of the returned object.
Changes in pomp version 3.4.2
Fix bug in
bsmc2
arising when some particles have zero likelihood. This is Issue #152. Thanks to Hola-Kwame Adrakey for reporting it.
Changes in pomp version 3.4.1
It is now possible to call
partrans
directly on an objective function, such as those created bytraj_objfun
,nlf_objfun
,probe_objfun
, orspect_objfun
.
Changes in pomp version 3.3.1
The package manual and help pages has been streamlined.
The
mif2
diagnostic plots have been made more flexible.Some deprecated functions have been removed. These include
cond.loglik
,sliceDesign
,runifDesign
,sobolDesign
, andprofileDesign
, which have been replaced bycode.logLik
,slice_design
,runif_design
,sobol_design
, andprofile_design
, respectively.
Changes in pomp version 3.3.0
A bug in
rgammawn
has been fixed.
Changes in pomp version 3.2.0
The pomp C API has been expanded a bit, to facilitate other packages' linking to the pomp library.
Changes in pomp version 3.1.1
The help pages have been improved, with an description of the overall package structure and pages on basic model components, elementary algorithms, and estimation algorithms.
A note to Windows users on avoiding certain C snippet compilation errors has been added to the help pages.
The behavior of
profile_design
has changed. Previously, the same random (or sub-random) sequence was used for all non-profile parameters. Now, for each profile slice, a distinct set of points is generated.-
profileDesign
,sliceDesign
,runifDesign
, andsobolDesign
have been deprecated and will be removed in a future release. They are replaced byprofile_design
,slice_design
,runif_design
, andsobol_design
, respectively.
Changes in pomp version 3.0.3
A new
saved.states
method allows one to extract the saved states from a particle filter computation.
Changes in pomp version 3.0.2
All instances of
cond.loglik
have been changed tocond.logLik
, to remove a common source of typographical error.
Changes in pomp version 3.0.1
A new sequential importance sampling algorithm has been implemented as
wpfilter
. This is a generalization of the algorithm inpfilter
in that it tracks the weights of the particles and allows the user to customize the resampling scheme.
Changes in pomp version 3.0.0
In
mif2
, the specification of particle numbers,Np
, has changed. WhenNp
is supplied as a function,Np(0)
is the requested number of particles at the beginning of the time series. The previous behavior was thatNp(1)
specified the requested number of particles. This behavior now matches that of the other particle filtering algorithms,pfilter
,bsmc2
, andpmcmc
.-
bsmc2
can now accept a variable number of particles, as do the other particle-filter based algorithmspfilter
,pmcmc
, andmif2
. The internal
systematic_resample
function now allows the user to specify the number of samples desired. Previously, these were always equal to the number of weights supplied.As promised from version 2.4.1, the
tol
andmax.fail
arguments have been removed completely from all particle-filtering algorithms, includingpfilter
,pmcmc
,bsmc2
, andmif2
. See the pomp news blog (https://kingaa.github.io/pomp/blog.html) for more information.The long-deprecated functions
onestep.dens
,onestep.sim
,discrete.time.sim
,euler.sim
,gillespie.sim
,gillespie.hl.sim
,conv.rec
, andvalues
have been removed. These have been replaced as followsonestep.dens
direct specification of dprocess
componentonestep.sim
onestep
discrete.time.sim
discrete_time
euler.sim
euler
gillespie.sim
gillespie
gillespie.hl.sim
gillespie_hl
conv.rec
traces
values
as.data.frame
oras(x,"data.frame")
Changes in pomp version 2.8.0
Trap error from non-finite likelihoods in
pmcmc
.New test of
pmcmc
codes in ‘tests/issue109.R’.Trap error from missing accumulator variables in
trajectory
.
Changes in pomp version 2.7.2
The Euler step size in
sir
is now adjustable.
Changes in pomp version 2.7.1
As promised, the default value of
tol
in the particle-filtering algorithmspfilter
,pmcmc
,mif2
, andbsmc2
has been changed to zero. A warning continues to be issued iftol
is set to anything other than 0. In a future release, the option to choose a nonzero tolerance will be removed entirely.In traces and diagnostic plotting methods, the
nfail
variable (tracking numbers of filtering failures) has been dropped.
Changes in pomp version 2.6.0
Another documentation bug was fixed.
Changes in pomp version 2.5.0
A documentation bug, identified by K. Hornik, was fixed.
Changes in pomp version 2.4.1
In anticipation of forthcoming algorithmic changes, the
tol
andmax.fail
arguments ofpfilter
,bsmc2
,pmcmc
, andmif2
are now deprecated. A warning is issued iftol
is set to anything other than 0. In a future release, these options will be removed entirely: the behavior will be similar to that obtained in the present release by settingtol=0
.A new example illustrating the
accumvars
argument has been provided.Internal computations now use increased precision for means, variances, and certain calculations with likelihoods.
A new help page,
pomp_examples
, lists the various example datasets and pomp objects provided with the package.
Changes in pomp version 2.3.0
The source codes underlying specification of basic model components via R functions has been reworked to remove reliance on deep
PROTECT
stacks. Thanks to Thomas Kalibera for showing the way!A bug in the documentation for
filter.traj
, pointed out by Pierre Jacob, has been fixed.
Changes in pomp version 2.2.2
The C API for pomp is no longer explained only in comments in the “pomp.h” header file. It is now described and explained in the new C API vignette.
Changes in pomp version 2.2.1
The new
pomp_cdir
global option allows one to set the directory holding C snippet files globally.A bug caused by re-use of the
cfile
argument has been fixed. Re-use now triggers an error, with an informative message.
Changes in pomp version 2.2.0
Some documentation improvements.
Changes in pomp version 2.1.2
A bug in
runifDesign
(and thereforeprofileDesign
whentype="runif"
) has been fixed.
Changes in pomp version 2.1.1
There are now
plot
methods for objects of classes ‘probe_match_objfun’ and ‘spect_match_objfun’.
Changes in pomp version 2.1.0
Released to CRAN.
Changes in pomp version 2.0.14
The
melt
function can now be applied to arbitrary ‘pomp’ objects.
Changes in pomp version 2.0.13
A new function,
runifDesign
allows construction of design matrices using samples from multivariate uniform distributions.The
profileDesign
function takes a new argument,type
, with two options:"sobol"
, the default, constructs the design usingsobolDesign
, as before."runif"
uses the newrunifDesign
function.
Changes in pomp version 2.0.12
A new example, comprising models and data for the 2014–2016 west African Ebola virus disease outbreak, is included. See
?ebola
for details.
Changes in pomp version 2.0.11
It is now possible to adjust the observation times and zero-time in the
sir
andsir2
examples. One can also change the default RNG seed used in generating the simulated data.The codes underlying the
rprocess
workhorse have been streamlined to obviate an unnecessary copy operation. As a result, the interface to therprocess
workhorse has been changed.The interface to the new
flow
workhorse has been changed to match.
Changes in pomp version 2.0.10
The new
flow
workhorse allows integration/iteration of the deterministic skeleton at arbitrary times from an arbitrary initial condition.
Changes in pomp version 2.0.9
When
coef
is applied to a ‘listie’, the result can be more usefullymelt
ed.The probe-names returned by
probe.acf
andprobe.ccf
have changed.A bug, whereby the ‘coef’ method from other packages was masked, has been fixed.
Errors in the documentation have been corrected.
Changes in pomp version 2.0.8
There is a new
as.data.frame
method for ‘listie’s.Some problems with the exposure of internal objects for package developers have been corrected.
Changes in pomp version 2.0.7
The magrittr pipe operator,
%>%
, is now re-exported.
Changes in pomp version 2.0.6
More examples have been included.
Minor changes to error messages.
Changes in pomp version 2.0.5
The names of the ‘objfun’ methods have been changed to
probe_objfun
,spect_objfun
,traj_objfun
, andnlf_objfun
.
Changes in pomp version 2.0.4
The package directory structure has been modified to include a new ‘examples’ directory, which holds the examples used in the documentation.
The names of the ‘objfun’ methods have been shortened.
Changes in pomp version 2.0.3
It is now permissible that the sequence of observation times (and sequences of covariate times) be merely non-decreasing sequences. Previously, pomp insisted on strictly increasing time sequences.
The default cooling schedule (
cooling.type
) inmif2
is now “geometric”, in contrast to “hyperbolic”, as before.
Changes in pomp version 2.0.2
-
pomp now requires at least R version 3.5.
Changes in pomp version 2.0.1
The
zeronames
argument has been renamedaccumvars
.
Changes in pomp version 2.0.0
In the course of ordinary work, one should never need to interact with the low-level
pomp
constructor function now.The
pompExample
function has been dropped. The various examples are now included as regular functions that construct ‘pomp’ objects.The
bbs
example has been dropped.One can now modify basic model components at (almost) any stage in a pipeline of pomp computations. All the top-level inference functions, plus
probe
,pfilter
, andsimulate
now allow modification of basic model components.The
rprocess
pluginsonestep.sim
,discrete.time.sim
,euler.sim
,gillespie.sim
, andgillespie.hl.sim
have been renamedonestep
,discrete_time
,euler
,gillespie
, andgillespie_hl
, respectively. The old usage is available but deprecated.One can now call
simulate
on a model without reference to any data.The covariate table facility (
covariate_table
) now supports piecewise constant interpolation.The syntax with which one includes covariates via
covariate_table
has become more flexible. Specifically, the arguments tocovariate_table
are now evaluated sequentially, so that later ones can depend on earlier ones. Once evaluated, the covariates are bound column-wise into a single data frame.The manner in which one writes R functions to specify basic model components has been totally changed. Before, one wrote functions that took specific arguments such as 'x', 'params', and 'covars'. Now, one writes such functions with any or all state variables, observables, covariates, and/or time as arguments.
-
simulate
now returns more informative results when simulations from multiple parameter sets are simultaneously computed. Specifically, ifparams
has column names, these are used to identify the resulting simulations. Thus whenformat = "pomps"
(the default), the names of the resulting list will be constructed from the column names ofparams
. Likewise, whenformat = "arrays"
, the resulting arrays will have informative column names; Whenformat = "data.frame"
, the identifier variable will make use of the column names. The basic particle filter,
pfilter
, has a simpler mode of operation:params
should be a single parameter set only. That is, it is no longer possible to pass a matrix of parameters topfilter
.The Liu-West algorithm,
bsmc2
, has a simpler mode of operation:params
should be a single parameter set only: the SMC particles are drawn fromrprior
.The
as.data.frame
,states
, andobs
options forsimulate
have been done away with. One now chooses the format of the returned simulations via theformat
argument. When one simulates at multiple parameter sets, list names, arraydimnames
, or identifying variables in the output data frame help to correlate simulations with parameter sets.The
as.data.frame
argument totrajectory
has been removed in favor of a newformat
argument that allows one to choose between receiving the results in the form of an"array"
or a"data.frame"
.All the basic model components now have defaults.
The default
rinit
behavior remains as it was: it assumes the initial state distribution is concentrated at a point mass determined by parameters with “.0
” or “_0
” suffices.The default process model is “missing”: calls to
dprocess
andrprocess
will result in missing values (NA
).The default measurement model is “missing” as well.
The default prior is flat and improper: all calls to the default
dprior
result in1
(0
iflog = TRUE
, and all calls torprior
result inNA
.The default skeleton is missing.
The default parameter transformations remain the identity.
The
transform
argument present in many pomp algorithms has been removed. Parameter transformations are now automatically performed when appropriate. The default parameter transformation remains the identity.The old
probe.match
andtraj.match
functions have been removed. The new approach to parameter estimation based on numerical optimization involves constructing stateful objective functions. There are now three of these:traj.match.objfun
,probe.match.objfun
, andspect.match.objfun
.The documentation system has been thoroughly overhauled.
In
skeleton
, thet
argument has been replaced bytimes
, to make this uniform with the other workhorse functions.Covariates are now included using the
covariate_table
construction, passed via thecovar
argument topomp
. Thetcovar
argument is no longer used.Instead of providing separate
fromEstimationScale
andtoEstimationScale
arguments topomp
, one now provides a singlepartrans
argument constructed by means of the new functionparameter_trans
. In addition, references to “fromEstimationScale” and “toEstimationScale” have been abbreviated to “fromEst” and “toEst”, respectively, throughout the package.When writing C snippets for parameter transformations, the notation
T_p
refers to the value ofp
on the estimation scale. Thus, if parameteralpha
is to be log-transformed for estimation, thetoEst
snippet would contain the lineT_alpha = log(alpha);
and thefromEst
snippet would includealpha = exp(T_alpha);
.The
parameter_trans
function now takes optional argumentslog
,logit
, andbarycentric
, with names of parameters to be log-, logit-, and log-barycentric transformed, respectively.The
measurement.model
argument topomp
has been removed. It is now necessary to specify the measurement model directly usingrmeasure
and/ordmeasure
.The
initializer
is no longer referred to by that name. It is nowrinit
, since it draws from the distribution of the latent state at the initial time. The low-level functioninit.state
has been replaced byrinit
.The
conv.rec
method has been deprecated, replaced by the newtraces
method.The
euler.sir
example has been renamedsir
.The
gillespie.sir
example has been renamedsir2
.The
covmat
method can now be applied to aprobed.pomp
object. It returns the estimated covariance matrix of the probes.When
trajectory
calls on deSolve routines to numerically integrate a model vectorfield, more informative error messages are generated, and diagnostics are printed whenverbose = TRUE
.The
show
method applied to pomp objects is now quite terse. To obtain detailed information, thespy
method is available.The
spy
method now returns the information formerly displayed in a call toshow
, in addition to the C files associated with any C snippets.The use of
$
methods to access the slots of pomp S4 objects has now been removed. These classes are no longer exported (for the most part).The
transform
argument to theprobe.quantile
function has been removed (as it is entirely redundant!).The old and deprecated
mif
andbsmc
methods have been removed.
Changes in pomp version 1.18.8
The
obs
andstates
arguments ofsimulate
are deprecated and will be removed in a forthcoming release.The C-level interface for
apply_probe_sims
has changed: the new argumentrho
must be the “package:pomp” environment.
Changes in pomp version 1.18.7
A bug in
pfilter
(and hencepmcmc
) whenfilter.traj=TRUE
was discovered and fixed. In the buggy version, the filter trajectory was sampled with incorrect weights.Timo Voipio contributed a pull request that traps the error resulting from
sannbox
being passed acandidate.dist
that is not a function.
Changes in pomp version 1.18.6
-
show
andprint
methods have been made very terse: they simply report on the class of the object in question. To see more detailed information about a ‘pomp’ object, usespy
. New (unexported)
solibs<-
method allows developers of packages extending pomp to incorporate C snippets into object that contain ‘pomp’ objects.
Changes in pomp version 1.18.5
New
verbose
option intrajectory
gives diagnostic information from deSolve integration routines. Of course, this is pertinent only if the model skeleton is a vectorfield.Bug in the display of
istate
whendeSolve::ode
errors has been fixed.
Changes in pomp version 1.18.4
New
hitch
function facilitates construction ofpomp.fun
objects from R functions, C snippets, and links to external libraries.Native routines needed to access
pomp.fun
objects in C are now registered for use by other packages.-
coef(object) <- NULL
, whereobject
is of class ‘pomp’, erases any parameters stored inobject
.
Changes in pomp version 1.18.3
The
euler.sir
,gillespie.sir
,blowflies
,ricker
,bbs
,dacca
, andrw2
examples are all now implemented using C snippets.A new ‘pomp_defines.h’ header is provided for developers of packages that depend on pomp.
Internally,
rprocess
is no longer stored as an R function. The use of plugins is now required for the implementation of therprocess
component. Currently, there are five plugins available:onestep.sim
,discrete.time.sim
,euler.sim
,gillespie.sim
, andgillespie.hl.sim
.The new
hitch
function automates the building and linking of C snippets and R functions intopomp.fun
s.The handling of
dprocess
has changed. Theonestep.dens
plugin is now deprecated and will soon be removed.dprocess
is now specified directly using either a C snippet or an R function, in much the same way that, for example,dmeasure
is specified.The old ‘forward’ and ‘inverse’ options for the
dir
argument ofpartrans
have been dispensed with. The much more descriptive options ‘toEstimationScale’ and ‘fromEstimationScale’ remain.
Changes in pomp version 1.18.2
The C snippet compiling facilities have been thoroughly reworked.
Changes in pomp version 1.18.1
-
onestep.dens
can now take aCsnippet
specification of the process-model transition density.
Changes in pomp version 1.17.4
All inference functions can now take a list in the
start
orparams
arguments.
Changes in pomp version 1.17.3
When altering parameters in a call to
probe
on aprobed.pomp
object, the new parameters were ignored. This bug has been fixed.-
mif2
,pfilter
,probe
,probe.match
, andprobe.match.objfun
can now gracefully handle a list in thestart
orparams
arguments. New
spy
method displays the C snippet file(s) associated with apomp
object.
Changes in pomp version 1.17.2
The long-deprecated
seed
argument tobsmc
andbsmc2
has been removed.
Changes in pomp version 1.17.1
More work to obviate warnings generated by T. Kalibera's latest
rchk
checks.
Changes in pomp version 1.16.4
Messages from the compiler resulting from attempts to compile C snippets are now suppressed on all platforms except in case of failure. Setting
verbose=TRUE
forces display of these messages.Source codes have been revisited to obviate warnings generated by T. Kalibera's latest
rchk
checks.
Changes in pomp version 1.16.3
Fix a minor bugs in
spect
.Improve test coverage.
-
spect.match
has been refactored, with small changes to the interface. -
pomp
now handles specification of workhorse functions (i.e.,rprocess
,dprocess
,rmeasure
,dmeasure
,initializer
,skeleton
,rprior
,dprior
,fromEstimationScale
,toEstimationScale
) asNULL
more consistently.
Changes in pomp version 1.16.2
Fix bug in systematic resampler when
np != nw
.Fix bug in
conv.rec
when only one parameter is requested.Correct and improve documentation for
conv.rec
.-
coef<-
can now take a list of parameters: it simply appliesunlist
to turn the list into a numeric vector. Simplify
conv.rec
.
Changes in pomp version 1.16.1
In
pfilter
,Np
is now a required argument unlessparams
is specified as a matrix of particles. Thanks to Carles Breto for catching the unintended behavior.
Changes in pomp version 1.15.4
The bug reported in issue #57 has been fixed.
A bug arising when building pomp objects with Csnippets in certain Windows configurations has been fixed.
Changes in pomp version 1.15.3
Issue #56, in which covariate names are discarded when coercing a pomp object to a data frame, has been fixed. Thanks to Eamon O'Dea for reporting this bug.
More informative error messages are given when
bake
orfreeze
returnNULL
. In such a case, these functions now generate a warning and return a character-string message, with attributes.Improve documentation of
traj.match
, in response to Issue #57.
Changes in pomp version 1.15.2
A new facility,
gillespie.hl.sim
, allows implementation of Gillespie's algorithm for exact stochastic simulation of continuous-time Markov processes via a streamlined C snippet interface.The old facility for Gillespie simulations,
gillespie.sim
, now accepts C snippets for faster calculation of event rates.The old
skeleton.type
andskelmap.delta.t
arguments of thepomp
constructor, which have generated error messages since version 1.9.2, have been done away with completely.
Changes in pomp version 1.15.1
Eamon O'Dea identified a bug (Issue #54) occurring when
pomp
is called with a single-row data frame. This has been fixed using his contributed patch.
Changes in pomp version 1.14.2
A new
dmultinom
function is provided in ‘pomp.h’. This provides the probability distribution function for the multinomial distribution.
Changes in pomp version 1.14.1
Improved tests.
Fixed bugs emerging in R-devel due to refactoring of underlying R source code.
Changes in pomp version 1.13.4
Eamon O'Dea pointed out that the bug in
gillespie.sim
, addressed by version 1.12.3, was not completely fixed. This problem has now been resolved. In addition, covariates are now guaranteed to be evaluated at least once per observation interval.The new argument
hmax
togillespie.sim
allows the user to specify the maximum step size that will be taken before covariates are evaluated.The K-leap method has been removed. Its accuracy is low on test problems and it is difficult to make it compatible with the pomp paradigm. In particular, it does not readily accomodate time-varying covariates in a sensible way, nor does it lend itself to simulations at pre-specified times.
Changes in pomp version 1.13.3
-
bspline.basis
andperiodic.bspline.basis
now take an optional non-negative integer argumentderiv
. Whenderiv > 0
, the order-deriv
derivative of each basis function is returned. A new function,
periodic_bspline_basis_eval_deriv
, has been added to the C API. This function gives access, at the C level, to evaluation of the derivatives of periodic B-splines.
Changes in pomp version 1.13.2
Upgrade license to GPL version 3.
Changes in pomp version 1.13.1
Fix buffer-flow problem identified by
valgrind
.
Changes in pomp version 1.12.3
Fix bug found by Eamon O'Dea in
gillespie.sim
.Fix bug in
onestep.sim
.
Changes in pomp version 1.12.2
The Ricker example has a new parameter,
c
, which scales the (unobserved) population size. By default,c=1
, so that the behavior is as before.
Changes in pomp version 1.12.1
Improved error trapping.
Changes in pomp version 1.11.3
Some changes have been made to the use of random numbers in some codes (the stochastic simulations in
gillespie.sim
and the resampling in the particle-filtering codes). These changes will mean that calculations performed before will not be reproducible in perfect detail, but will be statistically equivalent.Minor changes in C codes to obviate warnings generated by T. Kalibera's
rchk
checks.
Changes in pomp version 1.11.2
The rare and fascinating “Bug of St. Patrick” has been fixed. Thanks to Carles Breto for finding the bug and helping to isolate it.
More informative error messages are generated when
dmeasure
returns a non-finite likelihood in any of the particle filtering codes.
Changes in pomp version 1.11.1
Some streamlining of codes.
Changes in pomp version 1.10.6
Callable routines are now registered.
Changes in pomp version 1.10.5
More verbosity is provided with
verbose=TRUE
.
Changes in pomp version 1.10.4
The functions
bake
,stew
, andfreeze
now preserve information about the system time used in computation as an attribute of the returned object. In addition, these functions store information about the RNG settings.Some documentation improvements, courtesy of Sebastian Funk.
Changes in pomp version 1.10.3
Documentation improvements.
Changes in pomp version 1.10.2
More verbosity in building of Csnippets.
Changes in pomp version 1.10.1
Better handling of
pomp
'stimes
argument.
Changes in pomp version 1.9.3
New
logLik
method forpmcmcList
objects.
Changes in pomp version 1.9.2
The deprecated options to supply a matrix or numeric vector as the
data
argument topomp
have been removed and doing so now generates an error. The first argument topomp
,data
, must now be provided either as a data frame or as an object of classpomp
. See?pomp
for details.The deprecated
skeleton.type
andskelmap.delta.t
arguments topomp
have been removed. Using these arguments will now generate an error message. The correct method of specifying the deterministic skeleton is now explained in the help pages (?pomp
).
Changes in pomp version 1.9.1
Source code fixes to eliminate warnings during compilation.
Improved test suite.
Minor documentation correction.
Changes in pomp version 1.8.11
During particle filtering, when non-finite
dmeasure
values are generated, an error is generated and the offending parameters, states, data, and time are reported. In methods that use parameter transformations (transform=TRUE
), the reported parameters were on the transformed (estimation) scale, which can be confusing. The error message now gives the parameters on the natural (model) scale.Modified ‘src/ssa.c’ to fix UBSAN errors.
Changes in pomp version 1.8.10
-
C codes have been modified to take advantage of changes to the R API (R SVN REVISION 71180).
Changes in pomp version 1.8.9
A bug in
mif2
arising when filtering fails in the final timestep has been fixed. In this case, when all particles are deemed inconsistent with the data, we use an unweighted mean (with a warning) in place of the default weighted mean.
Changes in pomp version 1.7.7
The stochastic simulation algorithms underlying the Gillespie and K-leap methods (
gillespie.sim
,kleap.sim
) have been refactored in C. In addition, errors due to accumulation of round-off error have been repaired. The result will be more accurate, but somewhat slower when there are a very large number of reactions.Further improvements to the test suite. Code coverage is now 100%.
Changes in pomp version 1.7.6
In Sobol' sequence generation (
sobolDesign
,profileDesign
), we now use the suggestion of Joe & Kuo (2003) to choose a better sequence start point.
Changes in pomp version 1.7.5
The generation of Sobol' sequences (
sobolDesign
,profileDesign
) is now performed using the NLopt routines written in C instead of the original TOMS 659 FORTRAN codes. The copyright notices are located in the source code.Improved testing suite.
Made
rw.sd
even more robust via newsafecall
class.
Changes in pomp version 1.7.4
Refactored
rw.sd
for greater stability.
Changes in pomp version 1.7.3
Fixed a bug in
plot
onpomp
objects with more than 10 variables to plot.Added the
shlib.args
option topomp
, allowing arbitrary options to be passed to the compiler when C snippets are used.Modified the “dacca” example to use C snippets.
Refactored
rw.sd
for greater stability.Made Sobol' codes more robust.
Changes in pomp version 1.7.2
Modified the demos to use best practices.
Changes in pomp version 1.7.1
Improved the test suite.
Changes in pomp version 1.6.7
Improve the test suite and error messages.
Replace the automatic “verbose” setting during compilation under Windows. This was dropped in version 1.6.4.3, when it no longer appeared to be necessary. Apparently, it is, at least under some Windows OS/R combinations.
Changes in pomp version 1.6.6
Fix
probe
bug introduced in version 1.6.5.9.
Changes in pomp version 1.6.5
Several obscure bug-fixes.
Improvement of the test suite and of error messages.
Changes in pomp version 1.6.4
Remove the automatic “verbose” setting during compilation under Windows. This was introduced in version 0.63-6 as a workaround. It appears no longer to be necessary.
Improve the
show
method forpomp
objects.Change the
print
method forpomp
objects.Improve messages for plugin errors.
Some internal refactoring.
Changes in pomp version 1.6.3
Improve error messages for missing basic components.
Changes in pomp version 1.6.2
Fix typos in one of the help-page examples.
Add some references.
Changes in pomp version 1.6.1
Major revisions to the package help pages and manual. The separate documentation on C snippets and process-model plugins has been merged into the main
pomp
help page, which now has comprehensive instructions on building apomp
object.Update to the package help page.
Changes in pomp version 1.5.10
The matrix and numeric vector options for the
data
argument topomp
are now deprecated and will be removed in a future release. In calls topomp
,data
should be either a data frame or apomp
-class object.Many more errors are now trapped and reported with more informative messages.
Changes in pomp version 1.5.9
Redundant
bootstrap
argument tonlf
has been removed.More improvements to the testing suite.
Changes in pomp version 1.5.8
Major improvements to error and warning handling.
Fix bugs that might arise in rare circumstances in
simulate
andeulermultinom
functions.More improvements to the testing suite.
Changes in pomp version 1.5.7
Many improvements to the test suite.
Fixed a bug in
pompExample
.
Changes in pomp version 1.5.6
Major overhaul of error handling system and error messages.
There is a new plugin for
rprocess
calledkleap.sim
. This implements the so-called “K-leap method” of Cai & Xu (2007).
Changes in pomp version 1.5.5
Many improvements to the test suite.
Changes in pomp version 1.5.4
More improvements to the test suite.
Fixed bugs found as a result of improved testing.
In
mif2
, it is now required thatNmif>0
.
Changes in pomp version 1.5.3
More improvements to the test suite.
Fixed bugs found as a result of improved testing.
Fixed bug in specification of skeleton in pompExamples.
Changes in pomp version 1.5.2
Improved the test suite to obtain better code coverage.
-
pomp
is no longer an S4 method. Its new status as an ordinary function abbreviates and consolidates the code and simplifies the documentation. Some minor improvements to error messages.
A bug restricting access in C code to
userdata
has been fixed.Fixed bug in
pompExample(...,show=TRUE)
.
Changes in pomp version 1.5.1
New methods for the Ensemble Kalman Filter (
enkf
) and Ensemble Adjustment Kalman Filter (eakf
) have been added to the package.
Changes in pomp version 1.4.9
A better interface for specifying a model's deterministic skeleton is provided. One specifies
skeleton=map(f,delta.t)
for a discrete-time skeleton (a map) andskeleton=vectorfield(f)
for a continuous-time skeleton (a vectorfield). The old argumentsskeleton.type
andskelmap.delta.t
are deprecated and will be removed in a future release.The
pfilter
help page has been improved. Specifically, the discussion of filtering failures is better.Miscellaneous improvements to the help pages.
Changes in pomp version 1.4.8
-
mif2
no longer computes filter means.
Changes in pomp version 1.4.7
The new argument
show
ofpompExamples
allows one to display the example code instead of executing it.The minimum version of R supported is now 3.1.2.
Changes in pomp version 1.4.6
-
init.state
now has the optional argumentnsim
. Using this, one can request multiple initial state vectors per parameter vector. -
pfilter
now uses less memory when it is asked to run at a single point in parameter space.
Changes in pomp version 1.4.5
A bug associated with the
rw.sd
argument tomif2
on Windows platforms has been fixed.
Changes in pomp version 1.4.4
The
method="mif2"
option tomif
has been removed. Usemif2
instead.
Changes in pomp version 1.4.3
The
particles
method (rarely if ever used), has been removed to streamline themif
codes.
Changes in pomp version 1.4.2
The
method="mif2"
option tomif
is now deprecated. It will be removed in a future release. Usemif2
instead.
Changes in pomp version 1.4.1
Following the publication of the Journal of Statistical Software announcement paper, the ‘CITATION’ has been modified to include the appropriate citation. Please use
citation("pomp")
to learn about how to cite pomp correctly.
Changes in pomp version 1.3.4
-
conv.rec
on the results of amif2
computation now play nicely withreshape2::melt
.
Changes in pomp version 1.3.3
It is now possible to direct
pomp
to placeCsnippet
codes in a specified location. This is accomplished via the newcdir
andcfile
arguments topomp
.
Changes in pomp version 1.3.2
When several concurrent R processes attempt to simultaneously compile
Csnipet
code from copies of apomp
object on the same machine, a race condition exists. This release fixes this by ensuring thatCsnippet
codes are written and compiled in a process-specific directory.
Changes in pomp version 1.3.1
This release removes several long-deprecated features.
The
seed
argument topfilter
has been removed. Usefreeze
to obtain similar functionality.The redundant
pars
argument tomif
has been removed. Names of parameters to be estimated are taken from the nonzero entries ofrw.sd
. Initial-value parameters are designated using theivps
argument.The old
parameter.transform
andparameter.inv.transform
arguments topomp
have been removed. They were superseded by the argumentsfromEstimationScale
andtoEstimationScale
in version 0.65-1.
Changes in pomp version 1.2.2
Some internal changes to
mif2
by C. Breto, in support of panel methods under development.
Changes in pomp version 1.2.1
Bug fix: for POMPs with a scalar state variable, the
filter.mean
,pred.mean
, andpred.var
methods dropped the singleton dimension of the respective arrays, in conflict with the assumptions of certainplot
methods. This resolves issue #3.
Changes in pomp version 1.1.4
Documentation improvements. Specifically, the details of how step-sizes are chosen in the
euler.sim
,discrete.time.sim
, andonestep.sim
plugins has been spelled out explicitly. Also, incompleteness in thesannbox
documentation has been corrected.
Changes in pomp version 1.1.3
Provide
filter.traj
method forpmcmcList
objects.
Changes in pomp version 1.1.2
Fixed a bug in
mif
andmif2
whencooling.type="hyperbolic"
andcooling.fraction.50=1
.