[Dream-commits] r17 - in pkg: . R demo man

[noreply at r-forge.r-project.org]

Author: josephguillaume
Date: 2010-03-03 08:48:11 +0100 (Wed, 03 Mar 2010)
New Revision: 17

Added:
   pkg/man/coef.dream.Rd
Modified:
   pkg/NAMESPACE
   pkg/R/predict.dream.R
   pkg/demo/00Index
   pkg/man/predict.dream.Rd
Log:
Updated help files to pass check

Modified: pkg/NAMESPACE
===================================================================
--- pkg/NAMESPACE	2010-03-03 04:55:18 UTC (rev 16)
+++ pkg/NAMESPACE	2010-03-03 07:48:11 UTC (rev 17)
@@ -1,4 +1,5 @@
 import(coda)
+import(stats)
 
 export(dream)
 export(CovInit)

Modified: pkg/R/predict.dream.R
===================================================================
--- pkg/R/predict.dream.R	2010-03-03 04:55:18 UTC (rev 16)
+++ pkg/R/predict.dream.R	2010-03-03 07:48:11 UTC (rev 17)
@@ -5,7 +5,7 @@
 ##' @return a dream object with approximately the requested number of function evaluations
 ## TODO. extra parameters to set in control?
 ## TODO: does not seem to yield stationary distribution?
-simulate.dream <- function(object,nsim=1000,seed=NULL){
+simulate.dream <- function(object,nsim=1000,seed=NULL,...){
   
   ## Generate more results from converged chains
   object$control$REPORT <- 0
@@ -37,7 +37,7 @@
 ##' @return  data frame with each column corresponding to a returned vector
 predict.dream <- function(object,newdata=NULL,
                           method="maxLik",level=0.99,
-                          last.prop=0.5,use.thinned=TRUE
+                          last.prop=0.5,use.thinned=TRUE,...
                           ){
 
   ## Check and initialise parameters

Modified: pkg/demo/00Index
===================================================================
--- pkg/demo/00Index	2010-03-03 04:55:18 UTC (rev 16)
+++ pkg/demo/00Index	2010-03-03 07:48:11 UTC (rev 17)
@@ -1,2 +1,3 @@
-FME.nonlinear.model Nonlinear model calibration as with FME vignette.
-run_example1 n-dimensional banana shaped gaussian distribution. From Vrugt's matlab code.
+FME.nonlinear.model	Nonlinear model calibration as with FME vignette.
+run_example1		Function call for n-dimensional banana shaped gaussian distribution. From Vrugt's matlab code.
+example1		Setup for n-dimensional banana shaped gaussian distribution. From Vrugt's matlab code.

Added: pkg/man/coef.dream.Rd
===================================================================
--- pkg/man/coef.dream.Rd	                        (rev 0)
+++ pkg/man/coef.dream.Rd	2010-03-03 07:48:11 UTC (rev 17)
@@ -0,0 +1,18 @@
+\name{coef.dream}
+\alias{coef.dream}
+\title{Extract parameter values from dream object}
+\usage{
+\S3method{coef}{dream}(object, last.prop=0.5, use.thinned=FALSE, method=c("maxLik", "mean", "median"),\dots)
+}
+\description{Extract parameter values using a choice of methods (or an
+  arbitrary function)}
+\value{named vector of parameter values}
+\arguments{
+  \item{object}{dream object}
+  \item{last.prop}{proportion of total sequence to use (0,1]
+    if 1, use whole sequence}
+  \item{use.thinned}{Use thinned MCMC chains}
+  \item{method}{either one of maxLik,mean,median or a function parameter
+    vector=f(dream object)}
+  \item{...}{Unused. Matches generic}
+}


Property changes on: pkg/man/coef.dream.Rd
___________________________________________________________________
Name: svn:executable
   + *

Modified: pkg/man/predict.dream.Rd
===================================================================
--- pkg/man/predict.dream.Rd	2010-03-03 04:55:18 UTC (rev 16)
+++ pkg/man/predict.dream.Rd	2010-03-03 07:48:11 UTC (rev 17)
@@ -1,14 +1,19 @@
 \name{predict.dream}
 \alias{predict.dream}
 \title{Predict values using dream object}
-\usage{predict.dream(object, newdata, method="maxLik", level=0.99, last.prop=0.5, use.thinned=TRUE)}
+\usage{
+\S3method{predict}{dream}(object, newdata=NULL,method="maxLik", level=0.99, last.prop=0.5, use.thinned=TRUE,\dots)
+}
 \description{
 Predict values using function calibrated by dream, optionally with new data,
 using various methods of summarising the posterior parameter and output distributions}
 \value{data frame with each column corresponding to a returned vector}
-\arguments{\item{object}{dream object}
-\item{newdata.}{new FUN.pars list. If NULL, use object's.}
-\item{method}{CI or a \code{method} of \code{\link\{coef.dream}}
-\item{level.}{Requested two-sided level of confidence. For CI method.}
-\item{last.prop}{Proportion of MCMC chains to keep}
-\item{use.thinned}{Whether to use existing thinned chains}}
+\arguments{
+  \item{object}{dream object}
+  \item{newdata}{new FUN.pars list. If NULL, use object's.}
+  \item{method}{CI or a \code{method} of \code{\link{coef.dream}}}
+  \item{level}{Requested two-sided level of confidence. For CI method.}
+  \item{last.prop}{Proportion of MCMC chains to keep}
+  \item{use.thinned}{Whether to use existing thinned chains}
+  \item{...}{Unused - matches generic}
+}