'coef' applied to 'listie's is now more meltable

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: pomp2
 Type: Package
 Title: Statistical Inference for Partially Observed Markov Processes
-Version: 2.0.8.1
-Date: 2018-12-18
+Version: 2.0.9.0
+Date: 2018-12-20
 Authors@R: c(person(given=c("Aaron","A."),family="King",
 		role=c("aut","cre"),email="[email protected]"),
 	  person(given=c("Edward","L."),family="Ionides",role=c("aut")),

R/coef.R

@@ -47,7 +47,9 @@ setMethod(
   "coef",
   signature=signature(object="listie"),
   definition=function(object, ...) {
-    do.call(cbind,lapply(object,coef))
+    x <- do.call(cbind,lapply(object,coef))
+    names(dimnames(x)) <- c("parameter",".id")
+    x
   }
 )
 

TODO.md

@@ -2,14 +2,14 @@
 
 ## For pomp 2:
 
+- for log-barycentric transformations, check that parameter are contiguous
+- do we need basic-component arguments at higher levels of the calling hierarchy for documentation purposes?  PROBABLY YES.
 - "guide to upgrading to **pomp** version 2"
 	- ~~discussion of stateful objective functions~~
 	- ~~include before and after codes for the examples from the JSS paper~~
-- for log-barycentric transformations, check that parameter are contiguous
 - more/better examples
 	- need examples of objective function methods
 - more toy models
-- do we need basic-component arguments at higher levels of the calling hierarchy for documentation purposes?  PROBABLY YES.
 - get the metaphors straight! "horses" vs "hitches" vs "workhorses" ("wagons"?)
 - manual pages on the organization of the package
     - estimation algorithms
@@ -36,7 +36,7 @@
 - `coef` returns data frame in some circumstances?
 - demonstration of Fisher information via `pfilter` on slice designs?
 - new `userdata` function to get void *?
-- `simulate` with `format="d"` should return the same thing as `simulate` %>% `as.data.frame`
+- ~~`simulate` with `format="d"` should return the same thing as `simulate` %>% `as.data.frame`~~
 
 - perhaps `dinit`?
 - easier interface for lists of probes in `probe`
@@ -47,19 +47,20 @@
 - MCMC proposals as pomp slots?
 - probes as pomp slots?
 - what does a generic `pomp.fun` interface look like?
+
 - should parameter transformations allow renaming of variables?
 	- would require attention to `rw.sd`, e.g.
+
 - weighted particle filter
 
 ## For pomp:
 
 - streamline the R snippets so that ... is unnecessary?
 	- probably not worth the trouble, since C snippets are so much faster
 - add `include` argument to `pomp`?
-- deprecate `obs` and `states` arguments in `simulate`
+- ~~deprecate `obs` and `states` arguments in `simulate`~~
 - graceful stopping for optimizers (at least for `nloptr`)
-- `pfilterList` object?
-- parameter transformations: put `transform` option into each estimation routine (`spect.match`)
+- ~~`pfilterList` object~~
 - trap errors for LAPACK
 - support for asymmetric MCMC proposals
 - one-point SCQL function for possible use in fitting initial conditions
@@ -71,8 +72,8 @@
 
 ## Documentation
 
-- FAQ needs update with new issues
-- Complete rewrite of manual for v. 2
+- FAQ needs update with new issues from @MarieAugerMethe
+- ~~Complete rewrite of manual for v. 2~~
 
 ## Helper packages
 

inst/NEWS

@@ -1,5 +1,17 @@
 _N_e_w_s _f_o_r _p_a_c_k_a_g_e '_p_o_m_p'
 
+_C_h_a_n_g_e_s _i_n '_p_o_m_p' _v_e_r_s_i_o_n _2._0._9:
+
+        • When ‘coef’ is applied to a ‘listie’, the result can be more
+          usefully ‘melt’ed.
+
+_C_h_a_n_g_e_s _i_n '_p_o_m_p' _v_e_r_s_i_o_n _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.
+
 _C_h_a_n_g_e_s _i_n '_p_o_m_p' _v_e_r_s_i_o_n _2._0._7:
 
         • The ‘magrittr’ pipe operator, ‘%>%’, is now re-exported.

inst/NEWS.Rd

@@ -1,5 +1,16 @@
 \name{NEWS}
 \title{News for package `pomp'}
+\section{Changes in \pkg{pomp} version 2.0.9}{
+  \itemize{
+    \item When \code{coef} is applied to a \sQuote{listie}, the result can be more usefully \code{melt}ed.
+  }
+}
+\section{Changes in \pkg{pomp} version 2.0.8}{
+  \itemize{
+    \item There is a new \code{as.data.frame} method for \sQuote{listie}s.
+    \item Some problems with the exposure of internal objects for package developers have been corrected.
+  }
+}
 \section{Changes in \pkg{pomp} version 2.0.7}{
   \itemize{
     \item The \pkg{magrittr} pipe operator, \code{\%>\%}, is now re-exported.