[Pomp-commits] r515 - in pkg: . inst inst/doc
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Wed Jun 8 17:35:39 CEST 2011
Author: kingaa
Date: 2011-06-08 17:35:39 +0200 (Wed, 08 Jun 2011)
New Revision: 515
Modified:
pkg/DESCRIPTION
pkg/inst/ChangeLog
pkg/inst/doc/intro_to_pomp.Rnw
pkg/inst/doc/intro_to_pomp.pdf
Log:
- remove the deterministic skeleton from the SIR-model pomp objects.
Modified: pkg/DESCRIPTION
===================================================================
--- pkg/DESCRIPTION 2011-06-07 08:23:04 UTC (rev 514)
+++ pkg/DESCRIPTION 2011-06-08 15:35:39 UTC (rev 515)
@@ -1,8 +1,8 @@
Package: pomp
Type: Package
Title: Statistical inference for partially observed Markov processes
-Version: 0.38-3
-Date: 2011-06-06
+Version: 0.38-4
+Date: 2011-06-08
Author: Aaron A. King, Edward L. Ionides, Carles Breto, Steve Ellner, Bruce Kendall, Helen Wearing, Matthew J. Ferrari, Michael Lavine, Daniel C. Reuman
Maintainer: Aaron A. King <kingaa at umich.edu>
URL: http://pomp.r-forge.r-project.org
Modified: pkg/inst/ChangeLog
===================================================================
--- pkg/inst/ChangeLog 2011-06-07 08:23:04 UTC (rev 514)
+++ pkg/inst/ChangeLog 2011-06-08 15:35:39 UTC (rev 515)
@@ -1,3 +1,26 @@
+2011-06-07 kingaa
+
+ * [r514] man/dacca.Rd, man/eulermultinom.Rd: - fix 'dacca.Rd' to
+ eliminate codoc warning
+ - provide ascii-only eqns for 'eulermultinom' help
+
+2011-06-06 kingaa
+
+ * [r513] man/gompertz.Rd: - fix Gompertz model documentation
+ * [r512] DESCRIPTION, inst/NEWS: - update NEWS file
+ * [r511] man/bsplines.Rd, man/dacca.Rd, man/eulermultinom.Rd,
+ man/gompertz.Rd, man/nlf.Rd, man/plugins.Rd, man/pomp-methods.Rd,
+ man/pomp.Rd, man/ricker.Rd: - documentation improvements
+
+2011-06-05 kingaa
+
+ * [r510] DESCRIPTION, R/pomp.R, inst/ChangeLog,
+ inst/doc/advanced_topics_in_pomp.pdf, inst/doc/intro_to_pomp.Rnw,
+ inst/doc/intro_to_pomp.pdf, src/pfilter.c, src/tsir.c: - remove
+ unused variables in C codes (per build warning)
+ - fix errors in 'intro to pomp' vignette
+ - minor change to some warnings in 'pomp'
+
2011-06-03 kingaa
* [r509] DESCRIPTION, inst/NEWS,
Modified: pkg/inst/doc/intro_to_pomp.Rnw
===================================================================
--- pkg/inst/doc/intro_to_pomp.Rnw 2011-06-07 08:23:04 UTC (rev 514)
+++ pkg/inst/doc/intro_to_pomp.Rnw 2011-06-08 15:35:39 UTC (rev 515)
@@ -1549,6 +1549,42 @@
step.fun=sir.proc.sim,
delta.t=1/52/20
),
+ measurement.model=reports~binom(size=cases,prob=exp(rho)),
+ initializer=function(params, t0, ic.pars, comp.names, ...){
+ x0 <- c(S=0,I=0,R=0,cases=0)
+ N <- exp(params["N"])
+ fracs <- exp(params[ic.pars])
+ x0[comp.names] <- round(N*fracs/sum(fracs))
+ x0
+ },
+ zeronames=c("cases"), # zero this variable after each obsvn.
+ ic.pars=c("S0","I0","R0"), # initial condition parameters
+ comp.names=c("S","I","R") # names of the compartments
+ ),
+ params=log(
+ c(
+ N=50000,
+ beta=60,gamma=8,mu=1/50,
+ rho=0.6,
+ S0=8/60,I0=0.002,R0=1-8/60-0.001
+ )
+ ),
+ seed=677573454L
+ ) -> sir
+@
+<<sir-pomp-def-with-skel,eval=F,echo=F>>=
+simulate(
+ pomp(
+ data=data.frame(
+ time=seq(1/52,15,by=1/52),
+ reports=NA
+ ),
+ times="time",
+ t0=0,
+ rprocess=euler.sim(
+ step.fun=sir.proc.sim,
+ delta.t=1/52/20
+ ),
skeleton.type="vectorfield",
skeleton=function(t,x,params,...){
N <- exp(params["N"]) # population size
@@ -1562,13 +1598,13 @@
x["I"]*c(gamma,mu), # exits from I
x["R"]*mu # exits from R
)
- ## now connect the compartments
+ ## now put the equations together:
c(
- terms[1]-terms[2]-terms[3],
- terms[2]-terms[4]-terms[5],
- terms[4]-terms[6],
## approx. number of new infections/week:
- (terms[2]-terms[4]-terms[5])*gamma/52
+ cases=unname((terms[2]-terms[4]-terms[5])*gamma/52),
+ S=unname(terms[1]-terms[2]-terms[3]), # dS/dt
+ I=unname(terms[2]-terms[4]-terms[5]), # dI/dt
+ R=unname(terms[4]-terms[6]) # dR/dt
)
},
measurement.model=reports~binom(size=cases,prob=exp(rho)),
Modified: pkg/inst/doc/intro_to_pomp.pdf
===================================================================
(Binary files differ)
More information about the pomp-commits
mailing list