typo

BayesianTools/R/classLikelihood.R

@@ -8,6 +8,7 @@
 #' @param sampler sampler
 #' @seealso \code{\link{likelihoodIidNormal}} \cr
 #' \code{\link{likelihoodAR1}} \cr
+#' @example /inst/examples/createLikelihoodHelp.R
 #' @export
 createLikelihood <- function(likelihood, names = NULL, parallel = F, catchDuplicates=T, 
  sampler = NULL, parallelOptions = NULL){

BayesianTools/inst/examples/VSEMHelp.R

@@ -1,5 +1,5 @@
 
- 
+
 ## This example shows how to run and calibrate the VSEM model 
 
 library(BayesianTools)

BayesianTools/inst/examples/marginalLikelihoodHelp.R

@@ -180,4 +180,3 @@ marginalLikelihood(out, method = "HM", numSamples = 500)$ln.ML - theory
 marginalLikelihood(out, method = "Bridge", numSamples = 500)$ln.ML - theory
 
 
-

BayesianTools/inst/examples/plotSensitivityHelp.R

@@ -1,5 +1,4 @@
-
 ll <- testDensityBanana
 bayesianSetup <- createBayesianSetup(likelihood = ll, lower = rep(-10, 2), upper = rep(10, 2))
-
 plotSensitivity(bayesianSetup)
+

BayesianTools/inst/examples/plotTimeSeriesResultsHelp.R

@@ -25,7 +25,7 @@ likelihood <- function(par, sum = TRUE){
  x[parSel] = par
  predicted <- VSEM(x[1:11], PAR) # replace here VSEM with your model 
  predicted[,1] = 1000 * predicted[,1] # this is just rescaling
- diff <- c(predicted[,1:4] - obs[,1:4]) # difference betweeno observed and predicted
+ diff <- c(predicted[,1:4] - obs[,1:4]) # difference between observed and predicted
  # univariate normal likelihood. Note that there is a parameter involved here that is fit
  llValues <- dnorm(diff, sd = x[12], log = TRUE) 
  if (sum == FALSE) return(llValues)
@@ -67,3 +67,4 @@ plotTimeSeriesResults(sampler = out, model = createPredictions, observed = obs[,
  error = createError, prior = TRUE, main = "Prior predictive")
 
 }
+

BayesianTools/inst/examples/proposalGeneratorHelp.R

@@ -59,3 +59,4 @@ testGenerator$returnProposal(testVector)
 x <- testGenerator$returnProposalMatrix(testMatrix)
 boxplot(x)
 table(x[,4])
+

BayesianTools/inst/examples/testLinearModel.R

@@ -1,3 +1,4 @@
+
 x = c(1,2)
 y = testLinearModel(x)
 plot(y)

BayesianTools/inst/examples/tracePlotHelp.R

@@ -1,3 +1,5 @@
+
+
 # set up and run the MCMC
 ll <- function(x) sum(dnorm(x, log = TRUE))
 setup <- createBayesianSetup(likelihood = ll, lower = c(-10, -10), upper = c(10,10))

BayesianTools/vignettes/BayesianTools.Rmd

@@ -307,7 +307,7 @@ The prior in the BayesianSetup consists of four parts
 - lower / upper boundaries
 - Additional info - best values, names of the parameters, ...
 
-These information can be passed by first creating an a extra object, via createPrior, or through the the createBayesianSetup function.
+These information can be passed by first creating an extra object, via createPrior, or through the the createBayesianSetup function.
 
 #### Creating priors
 
@@ -316,7 +316,7 @@ You have 5 options to create a prior
 - Do not set a prior - in this case, an infinite prior will be created
 - Set min/max values - a bounded flat prior and the corresponding sampling function will be created
 - Use one of the pre-definded priors, see ?createPrior for a list. One of the options here is to use a previous MCMC output as new prior. Pre-defined priors will usually come with a sampling function
-- Use a user-define prior, see ?createPrior
+- Use a user-defined prior, see ?createPrior
 - Create a prior from a previous MCMC sample
 
 #### Creating user-defined priors
@@ -568,7 +568,7 @@ MCMCs sample the posterior space by creating a chain in parameter space. While t
 
 An alternative to MCMCs are particle filters, aka Sequential Monte-Carlo (SMC) algorithms. See Hartig, F.; Calabrese, J. M.; Reineking, B.; Wiegand, T. & Huth, A. Statistical inference for stochastic simulation models - theory and application Ecol. Lett., 2011, 14, 816-827
 
-### Rejection samling
+### Rejection sampling
 
 The easiest option is to simply sample a large number of parameters and accept them according to their posterior value. This option can be emulated with the implemented SMC, setting iterations to 1.