Last updates 0.1.3

BayesianTools/DESCRIPTION

@@ -1,7 +1,7 @@
 Package: BayesianTools
 Title: General-Purpose MCMC and SMC Samplers and Tools for Bayesian Statistics
-Version: 0.1.2.1
-Date: 2017-05-28
+Version: 0.1.3
+Date: 2017-08-01
 Authors@R: c(person("Florian", "Hartig", email = "florian.hartig@biologie.uni-regensburg.de", role = c("aut", "cre")), person("Francesco", "Minunno", role = c("aut")), person("Stefan", " Paul", role = c("aut") ),  person("David", "Cameron", role = "ctb"))
 Author: Florian Hartig [aut, cre],
     Francesco Minunno [aut],

BayesianTools/NEWS

@@ -5,19 +5,14 @@ NOTE: for more news about the package, see https://github.com/florianhartig/Baye
 Changes
 
 - getSample extended to coda objects, so that all plot functions can also be used on coda objects 
-
-- getSample now corrects wrong inputs for thin and numSamples instead of throwing an error
-
 - plotTimeSeriesResults now includes the option to create residual plots from posterior predictive simulations, calling the DHARMa package
-
 - added new utitility functions sampleEquallySpaced and correctThin
 
-- added manual tests for getSample and automated tests for new utitlity functions
-
 Bugfixes
 
 - IMPORTANT: fixed a bug in createPriorDensity - results with createPriorDensity prior to this bugfix may be unrealiable
-
+- getSample now corrects wrong inputs for thin and numSamples instead of throwing an error
+- smaller bugfixes in plot functions
 
 # BayesianTools 0.1.2 
 

BayesianTools/inst/examples/VSEMHelp.R

@@ -57,6 +57,7 @@ out <- runMCMC(bayesianSetup = bayesianSetup, sampler = "DEzs", settings = setti
 
 plot(out)
 summary(out)
+marginalPlot(out, scale = T)
 gelmanDiagnostics(out) # should be below 1.05 for all parameters to demonstrate convergence 
 
 # Posterior predictive simulations
@@ -85,7 +86,11 @@ plotTimeSeriesResults(sampler = out, model = createPredictions, observed = refer
 
 
 ########################################################
-# demonstrating the updating of the prior 
+# Demonstrating the updating of the prior from old posterior
+# Note that it is usually more exact to rerun the MCMC 
+# with all (old and new) data, instead of updating the prior
+# because likely some information is lost when approximating the
+# Posterior by a multivariate normal 
 
 settings <- list(iterations = 5000, nrChains = 2)