2009-05-07  kingaa

	* [r121] inst/doc/compiled_code_in_pomp.pdf,
	  inst/doc/intro_to_pomp.pdf, inst/include/pomp.h,
	  src/eulermultinom.c, src/pomp.h, tests/sir.R,
	  tests/sir.Rout.save: streamline "reulermultinom" by moving
	  GetRNGstate() and PutRNGstate() outside of a level of looping. If
	  you want to use 'reulermultinom' at the C level, you must take
	  care that it is wrapped inside of the RNGstate functions. An
	  approx. 10-fold increase in speed for the euler.simulate function
	  is realized by this simple change.

2009-05-05  kingaa

	* [r120] inst/include/pomp.h: fix the pomp.h header comments

2009-05-01  kingaa

	* [r119] R/init.state-pomp.R: when params is missing, use params <-
	  coef(object)

2009-04-29  kingaa

	* [r118] src/pomp.h: fix comments
	* [r117] inst/ChangeLog, inst/doc/compiled_code_in_pomp.pdf,
	  inst/doc/intro_to_pomp.pdf:
	* [r116] DESCRIPTION, inst/include/pomp.h, man/euler.Rd,
	  src/euler.c, src/lookup_table.c, src/pomp.h, src/pomp_internal.h:
	  rearrange the headers
	  rename the prototypes for native routines used in Euler and
	  onestep plugins
	* [r115] src/bspline.c, src/dsobol.c:

2009-04-28  kingaa

	* [r114] inst/ChangeLog: version 0.23-4
	* [r113] data/rw2.rda, man/rw2.Rd, tests/rw2.R,
	  tests/rw2.Rout.save: add a new data()-accessible pomp object,
	  coding a very simple 2D random walk
	* [r112] DESCRIPTION, NAMESPACE, R/euler.R, data/euler.sir.rda,
	  data/ou2.rda, inst/ChangeLog, inst/doc/compiled_code_in_pomp.pdf,
	  inst/doc/intro_to_pomp.pdf, inst/examples/euler_sir.R,
	  inst/examples/logistic.R, inst/examples/rw2.R, man/euler.Rd,
	  src/euler.c, tests/logistic.R, tests/logistic.Rout.save,
	  tests/sir.R, tests/sir.Rout.save: add onestep plugin to
	  facilitate the construction of models for which explicit formulae
	  are available for one-step transitions.
	* [r111] inst/CITATION: fix citation info
	* [r110] src/dprocess.c, src/rprocess.c: explicitly cast 'times' to
	  double to avoid potential problems when R casts a sequence to
	  integer storage type
	* [r109] man/pomp-package.Rd:

2009-04-27  kingaa

	* [r108] inst/ChangeLog, inst/doc/Makefile,
	  inst/doc/compiled_code_in_pomp.pdf, inst/doc/intro_to_pomp.pdf:

2009-04-26  kingaa

	* [r107] DESCRIPTION, inst/ChangeLog: version 0.23-3
	* [r106] data/euler.sir.rda, data/ou2.rda: use R-2.8.1 to generate
	  the data/*.rda files
	* [r105] data/euler.sir.R, data/euler.sir.rda, data/ou2.R,
	  data/ou2.rda: It turns out not to work to use R files in the
	  'data' directory if these involve actual computations. It appears
	  that, during the package check process, the shared-object library
	  is loaded *after* the data files are read. Consequently, the
	  simulations used to generate the pomp objects in 'ou2.R' and
	  'euler.sir.R' will not work. As a temporary fix, I have replaced
	  these with binary files containing the objects. I do not yet have
	  a satisfactory way of including the code used to generate these
	  into the package yet.

2009-04-24  kingaa

	* [r104] R/trajectory-pomp.R, data/euler.sir.R, man/euler.sir.Rd,
	  tests/sir.R, tests/sir.Rout.save: put 'euler.sir' in as an
	  example pomp object obtainable via 'data(euler.sir)'
	  make 'trajectory' a bit more user-friendly
	* [r103] man/euler.Rd: remove the examples in the 'euler' and
	  'SSA.simulate' man pages: refer the user to the 'examples'
	  directory instead
	* [r102] DESCRIPTION, data/ou2.R, data/ou2.rda,
	  inst/doc/compiled_code_in_pomp.Rnw, tests/ou2-kalman.Rout.save,
	  tests/ou2-mif.Rout.save: replace the binary data 'ou2.rda' with R
	  code in 'ou2.R'

2009-04-21  kingaa

	* [r101] R/pomp.R: better error handling
	* [r100] R/pomp.R: time variable is not needed in the covar table

2009-04-17  kingaa

	* [r99] inst/ChangeLog:
	* [r98] DESCRIPTION, data/ou2.rda, inst/ChangeLog,
	  inst/doc/compiled_code_in_pomp.Rnw,
	  inst/doc/compiled_code_in_pomp.pdf, inst/doc/intro_to_pomp.pdf,
	  inst/examples/euler_sir.R, inst/examples/euler_sir.c,
	  inst/include/pomp.h, man/euler.Rd, src/euler.c, src/euler_sir.c,
	  src/ou2.c, src/pomp.h, src/rmeasure.c, tests/sir.R,
	  tests/sir.Rout.save: The 'euler' and 'rmeasure' evaluation codes
	  now call GetRNGstate() and PutRNGstate() when the user-supplied
	  function they will evaluate is a native routine. This means that
	  potentially many fewer calls to these expensive functions can be
	  made in, e.g., particle filtering or simulation of a large number
	  of replicates and consequently significant performance gains.

2009-04-16  kingaa

	* [r97] R/mif-methods.R, R/mif.R: use rowMeans and colMeans in
	  place of apply where appropriate
	* [r96] man/mif-class.Rd, man/mif-methods.Rd, man/mif.Rd: update
	  references for 'mif'
	* [r95] man/mif.Rd: fix order of arguments in documentation for
	  'mif'

2009-04-14  kingaa

	* [r94] inst/ChangeLog:
	* [r93] DESCRIPTION, R/mif.R, data/ou2.rda, inst/ChangeLog,
	  inst/doc/compiled_code_in_pomp.pdf, inst/doc/intro_to_pomp.Rnw,
	  inst/doc/intro_to_pomp.pdf, man/mif.Rd, tests/ou2-mif.R,
	  tests/ou2-mif.Rout.save, tests/sir.Rout.save: The parameter
	  'alg.pars' is now deprecated. The four algorithm parameters Np,
	  ic.lag, var.factor, cooling.factor are now specified directly as
	  required arguments to pomp. The 'alg.pars' argument will be
	  removed in a future release. Use of this argument now results in
	  a warning message.
	  

2009-04-13  kingaa

	* [r92] DESCRIPTION, data/ou2.rda, inst/ChangeLog,
	  inst/doc/compiled_code_in_pomp.pdf, inst/doc/intro_to_pomp.pdf,
	  tests/ou2-mif.Rout.save:
	* [r91] inst/ChangeLog:
	* [r90] R/mif.R, man/mif.Rd, tests/ou2-mif.R: rewrite mif methods
	  to use 'mif.internal'.
	  this obviates the need for a user-visible '.ndone' argument.
	  add some reporting when verbose=TRUE.
	  the 'particles' function can now be altered after the mif object
	  is created.
	  some new tests have been added to tests/ou2-mif.R
	* [r89] R/pfilter.R: a bit less reporting when verbose=TRUE

2009-04-11  kingaa

	* [r88] tests/ou2-mif.R, tests/ou2-mif.Rout.save: tests added for
	  'mif' and 'pfilter'
	* [r87] R/mif.R: modified in accordance with changes made in
	  revision 86.
	* [r86] R/pfilter-mif.R, R/pfilter.R, man/pfilter.Rd: put the guts
	  of the particle filter into 'pfilter.internal'.
	  this can be used by 'mif' directly and by methods for 'pfilter'.
	  this obviates the need for the argument .rw.sd in the
	  user-visible methods.

2009-04-10  kingaa

	* [r85] inst/ChangeLog:
	* [r84] DESCRIPTION:
	* [r83] DESCRIPTION, R/mif.R, R/pfilter.R, man/mif.Rd,
	  man/pfilter.Rd: The handling of the argument 'pars' and 'ivps' in
	  mif has been changed to eliminate redundancy and a possible
	  source of errors. Now 'pars' and 'ivps' are checked to make sure
	  that all names are in 'names(start)' and 'names(rw.sd)' and that
	  they are mutually disjoint. Further, all the entries in
	  'rw.sd[pars]' and 'rw.sd[ivps]' must be positive or an error is
	  generated. If 'rw.sd' has positive values in slots not in the
	  union of 'ivps' and 'pars', these are ignored and a warning is
	  generated. The 'random.walk.sd' slot of the resulting 'mif'
	  object contains only nonzero entries corresponding to the union
	  of 'pars' and 'ivps'.
	  
	  If 'pars' is not specified in the arguments, then it is taken to
	  be
	  'names(rw.sd)[!(names(rw.sd)%in%ivps)]'
	  
	  Thus it is no longer necessary to specify 'pars' explicitly,
	  provided one takes care that all and only nonzero entries in
	  'rw.sd' correspond to parameters that one wishes to estimate. One
	  can be a bit more careless with 'rw.sd' if one uses 'pars'. In
	  any case, entries in 'rw.sd' that are zero are ignored.
	  
	  The computation of prediction means and variances and filtering
	  means in 'pfilter' is handled differently. When used as a
	  toplevel function (i.e., when the .rw.sd parameter used
	  internally by 'mif' is unset), then the prediction means and
	  variances and the filtering means are only computed for the state
	  variables. This eliminates an annoying and unnecessary behavior
	  that arose when parameters were set to nonfinite values.
	  
	  When .rw.sd is set (as when 'mif' calls 'pfilter'), the
	  prediction means and variances are computed only for all state
	  variables, but for only those parameters which are undergoing the
	  MIF random walk. Filtering means are computed for all state
	  variables and parameters, since they are needed for the fixed-lag
	  smoothing on the initial-value parameters.
	* [r82] man/mif-class.Rd, man/mif-methods.Rd: fix some
	  documentation infelicities and bugs

2009-03-16  kingaa

	* [r81] inst/CITATION: fix Carles accent

2009-03-13  kingaa

	* [r80] man/pomp-methods.Rd: improve documentation for coef<-
	  method for pomp object

2009-03-11  kingaa

	* [r79] DESCRIPTION, inst/ChangeLog:
	* [r78] man/eulermultinom.Rd, src/eulermultinom.c: replace "euler
	  infelicity" Rprintf statements with warnings
	* [r77] inst/CITATION: add CITATION file

2009-02-03  kingaa

	* [r76] inst/ChangeLog:
	* [r75] DESCRIPTION, R/mif.R, man/mif.Rd, man/pomp-methods.Rd,
	  man/trajectory-pomp.Rd, src/euler_sir.c, src/sir.c: changes to
	  documentation and default arguments so as to bring these into
	  agreement. (discrepancies revealed by Kurt Hornik's extensive
	  codoc testing).

2009-02-01  kingaa

	* [r74] inst/NEWS, man/nlf.Rd:

2009-01-31  kingaa

	* [r73] DESCRIPTION:
	* [r72] inst/ChangeLog:
	* [r71] man/nlf.Rd: documentation fix

2009-01-28  kingaa

	* [r70] R/plot-pomp.R: remove default for the (unused) 'y' argument
	* [r69] man/nlf.Rd: fix typos

2009-01-27  kingaa

	* [r68] inst/ChangeLog:
	* [r67] DESCRIPTION, R/trajectory-pomp.R: 'odesolve' has been
	  replaced by 'deSolve'
	* [r66] inst/examples/euler_sir.R, inst/examples/euler_sir.c,
	  inst/examples/sir.R, inst/examples/sir.c: moved 'sir.*' examples
	  to 'euler_sir.*'
	* [r65] DESCRIPTION, R/plot-pomp.R, inst/ChangeLog,
	  inst/examples/sir.R, man/pomp-methods.Rd: add package 'subplex'
	  as dependency
	  run examples that compile native code only if
	  Sys.info()['sysname']=='Linux'. This replaces the previous test
	  that .Platform$OS.type=='unix', which resulted in errors under
	  some versions of MacOSX.
	* [r64] inst/ChangeLog:

2009-01-26  kingaa

	* [r62] inst/ChangeLog:

2009-01-25  kingaa

	* [r61] DESCRIPTION: minor documentation corrections
	* [r60] man/mif-class.Rd, man/mif-methods.Rd, man/mif.Rd:

2009-01-17  kingaa

	* [r59] inst/ChangeLog:
	* [r58] tests/ou2-nlf.Rout.save:
	* [r57] tests/ou2-mif.Rout.save, tests/ou2-nlf.Rout:
	* [r56] tests/ou2-nlf.R, tests/ou2-nlf.Rout.save:

2009-01-16  kingaa

	* [r55] DESCRIPTION: add functionality for parameter estimation via
	  nonlinear forecasting, method due to Stephen P. Ellner and Bruce
	  E. Kendall
	* [r54] NAMESPACE, R/mvnorm.R, R/nlf-funcs.R, R/nlf-guts.R,
	  R/nlf-lql.R, R/nlf-objfun.R, R/nlf.R, data/ou2.rda,
	  inst/ChangeLog, inst/doc/compiled_code_in_pomp.pdf,
	  inst/doc/intro_to_pomp.pdf, man/nlf.Rd, tests/ou2-nlf.R,
	  tests/ou2-nlf.Rout: add functionality for parameter estimation
	  via nonlinear forecasting, method due to Stephen P. Ellner and
	  Bruce E. Kendall
	* [r53] R/pfilter.R, man/pfilter.Rd: add 'verbose' option to
	  'pfilter'

2008-08-25  kingaa

	* [r52] data/ou2.rda, inst/ChangeLog,
	  inst/doc/compiled_code_in_pomp.pdf, inst/doc/intro_to_pomp.pdf:
	* [r51] R/mif.R, man/mif.Rd, tests/ou2-mif.Rout.save: error trap in
	  'mif': 'rw.sd' must have names
	* [r50] DESCRIPTION, inst/NEWS, man/euler.Rd: better documentation
	  in 'euler.Rd'

2008-08-19  kingaa

	* [r49] inst/NEWS:

2008-08-15  kingaa

	* [r48] data/ou2.rda, inst/ChangeLog,
	  inst/doc/compiled_code_in_pomp.pdf, inst/doc/intro_to_pomp.pdf:
	* [r47] tests/ou2-mif.Rout.save: fix bug in Euler step
	* [r46] tests/sir.R, tests/sir.Rout.save: improve test of SIR model
	* [r45] tests/logistic.Rout.save:
	* [r44] DESCRIPTION, src/euler.c: fix bug in Euler step

2008-08-14  kingaa

	* [r43] DESCRIPTION: make description more descriptive
	* [r42] inst/examples/sir.R: put trajectory computation into SIR
	  example

2008-08-13  kingaa

	* [r41] data/ou2.rda, inst/doc/compiled_code_in_pomp.pdf,
	  inst/doc/intro_to_pomp.pdf:
	* [r40] inst/ChangeLog:
	* [r39] inst/examples/sir.c: cosmetic
	* [r38] DESCRIPTION, NAMESPACE, R/aaa.R, R/pomp.R,
	  R/trajectory-pomp.R, data/ou2.rda, inst/examples/logistic.R,
	  inst/examples/sir.R, inst/examples/sir.c, man/mif-class.Rd,
	  man/pomp-class.Rd, man/pomp.Rd, man/skeleton-pomp.Rd,
	  man/trajectory-pomp.Rd, src/skeleton.c, tests/logistic.R,
	  tests/logistic.Rout.save, tests/sir.R, tests/sir.Rout.save: add
	  support for computing trajectories of the deterministic skeleton
	  the skeleton must now be specified when the pomp is created: if
	  the pomp is a discrete-time system, 'skeleton.map' is used to
	  specify its deterministic skeleton; if it is a continuous-time
	  system, 'skeleton.vectorfield' is used.
	  the new method 'trajectory' will, in the first case, iterate the
	  map to produce an itinerary and in the second case, it will call
	  'lsoda' from the now-required 'odesolve' package to compute the
	  trajectory.
	  some changes to the basic 'pomp' structure have been made to
	  facilitate all this.
	  specifically, a new slot, 'skeleton-type', distinguishes between
	  discrete-time and continuous-time systems

2008-08-01  kingaa

	* [r37] inst/ChangeLog:
	* [r36] inst/ChangeLog:
	* [r35] inst/examples/logistic.R: make dprocess depend on 'log'

2008-07-31  kingaa

	* [r34] inst/NEWS:
	* [r33] DESCRIPTION:
	* [r32] R/mif.R: improved error handling in mif
	* [r31] man/pomp.Rd: fix inaccuracy in documentation of rmeasure
	  and dmeasure arguments

2008-07-29  kingaa

	* [r29] inst/NEWS:
	* [r28] DESCRIPTION:
	* [r27] R/mif.R: bug fix in convergence record storage
	* [r26] R/mif.R: cosmetic

2008-07-26  kingaa

	* [r25] DESCRIPTION, inst/ChangeLog, inst/NEWS,
	  inst/doc/compiled_code_in_pomp.pdf, inst/doc/intro_to_pomp.Rnw,
	  inst/doc/intro_to_pomp.pdf, src/ou2.c: bug fix in
	  "intro_to_pomp.Rnw"

2008-07-25  kingaa

	* [r24] R/pfilter-mif.R:
	* [r23] inst/NEWS:
	* [r22] inst/NEWS:
	* [r21] data/ou2.rda, inst/ChangeLog,
	  inst/doc/compiled_code_in_pomp.pdf, inst/doc/intro_to_pomp.pdf:
	* [r20] tests/ou2-mif.Rout.save: no need to set rw.sd for
	  parameters that aren't estimated
	* [r19] inst/examples/logistic.R, inst/examples/rw2.R,
	  inst/examples/sir.R, tests/examples.R: add a test of the examples
	  in 'inst/examples' to the 'tests' directory
	  modify the sir.R example so that it attempts to build a
	  shared-object library only if .Platform$OS.type=='unix'
	* [r18] inst/doc/intro_to_pomp.Rnw, tests/ou2-mif.R: no need to
	  include zeros in 'rw.sd'
	* [r17] man/mif.Rd: improve documentation regarding arguments to
	  mif
	* [r16] R/aaa.R: remove a whole lot of stuff that somehow got in
	  from pomp.devel
	* [r15] inst/examples/sir.R: fix up R-function version
	* [r14] R/aaa.R: improve error handling: fix language and use
	  sQuote, dQuote where appropriate
	* [r13] DESCRIPTION:
	* [r12] R/compare.mif.R, R/dmeasure-pomp.R, R/dprocess-pomp.R,
	  R/euler.R, R/init.state-pomp.R, R/mif-methods.R, R/mif.R,
	  R/particles-mif.R, R/pfilter.R, R/plot-pomp.R, R/pomp-methods.R,
	  R/pomp.R, R/rmeasure-pomp.R, R/rprocess-pomp.R,
	  R/simulate-pomp.R, R/skeleton-pomp.R, R/sobol.R: improve error
	  handling: fix language and use sQuote, dQuote where appropriate
	* [r11] R/mif.R: trap as error parameters (including ivps) which
	  are to be estimated but which have zero rw.sd

2008-07-22  kingaa

	* [r10] inst/NEWS:
	* [r9] inst/ChangeLog:
	* [r8] inst/ChangeLog, inst/O_CHANGES:

2008-07-20  kingaa

	* [r7] DESCRIPTION, NAMESPACE, R/aaa.R, R/bsplines.R,
	  R/compare.mif.R, R/dmeasure-pomp.R, R/dprocess-pomp.R, R/euler.R,
	  R/eulermultinom.R, R/init.state-pomp.R, R/mif-methods.R, R/mif.R,
	  R/particles-mif.R, R/pfilter-mif.R, R/pfilter.R, R/plot-pomp.R,
	  R/pomp-methods.R, R/pomp.R, R/rmeasure-pomp.R, R/rprocess-pomp.R,
	  R/simulate-pomp.R, R/skeleton-pomp.R, R/sobol.R, inst,
	  inst/ChangeLog, inst/GPL, inst/LICENSE, inst/NEWS, inst/doc,
	  inst/doc/Makefile, inst/doc/compiled_code_in_pomp.Rnw,
	  inst/doc/compiled_code_in_pomp.pdf, inst/doc/intro_to_pomp.Rnw,
	  inst/doc/intro_to_pomp.pdf, inst/examples,
	  inst/examples/logistic.R, inst/examples/rw2.R,
	  inst/examples/sir.R, inst/examples/sir.c, inst/include,
	  inst/include/pomp.h, man/bsplines.Rd, man/dmeasure-pomp.Rd,
	  man/dprocess-pomp.Rd, man/euler.Rd, man/eulermultinom.Rd,
	  man/init.state-pomp.Rd, man/mif-class.Rd, man/mif-methods.Rd,
	  man/mif.Rd, man/ou2.Rd, man/particles-mif.Rd, man/pfilter.Rd,
	  man/pomp-class.Rd, man/pomp-methods.Rd, man/pomp-package.Rd,
	  man/pomp.Rd, man/rmeasure-pomp.Rd, man/rprocess-pomp.Rd,
	  man/simulate-pomp.Rd, man/skeleton-pomp.Rd, man/sobol.Rd, src,
	  src/bspline.c, src/dmeasure.c, src/dprocess.c, src/dsobol.c,
	  src/euler.c, src/eulermultinom.c, src/initstate.c,
	  src/lookup_table.c, src/ou2.c, src/pomp.h, src/pomp_fun.c,
	  src/pomp_internal.h, src/resample.c, src/rmeasure.c,
	  src/rprocess.c, src/sir.c, src/skeleton.c, src/sobol.f, tests,
	  tests/logistic.R, tests/logistic.Rout.save, tests/ou2-kalman.R,
	  tests/ou2-kalman.Rout.save, tests/ou2-mif.R,
	  tests/ou2-mif.Rout.save, tests/ou2-simulate.R,
	  tests/ou2-simulate.Rout.save, tests/rw2.R, tests/rw2.Rout.save,
	  tests/sir.R, tests/sir.Rout.save: initial revision
	* [r6] data/ou2.rda:
	* [r5] data:

2007-06-25  stefan7th

	* [r1] ., R, man: Email in Readme changed

