Produces a rank plot to analyse convergence of MCMC algorithm

plotConvergence(params)

Arguments

params

An instance of class bandleParams

Value

Returns the ranks of the number of outliers in each chain. The side effect returns rank plots. Number of rank plots is equal to the number of chains

Examples

## Generate some example data
library("pRolocdata")
data("tan2009r1")
set.seed(1)
tansim <- sim_dynamic(object = tan2009r1, 
                      numRep = 4L,
                      numDyn = 100L)
data <- tansim$lopitrep
control <- data[1:2]
treatment <- data[3:4]

## fit GP params
gpParams <- lapply(tansim$lopitrep, function(x) 
fitGPmaternPC(x, hyppar = matrix(c(0.5, 1, 100), nrow = 1)))

## run bandle
res <- bandle(objectCond1 = control,
              objectCond2 = treatment, 
              gpParams = gpParams,
              fcol = "markers",  
              numIter = 5L, 
              burnin = 1L, 
              thin = 2L,
              numChains = 2, 
              BPPARAM = SerialParam(RNGseed = 1),
              seed = 1)
#> 
  |                                                                            
  |                                                                      |   0%
  |                                                                            
  |==============                                                        |  20%
  |                                                                            
  |============================                                          |  40%
  |                                                                            
  |==========================================                            |  60%
  |                                                                            
  |                                                                      |   0%
  |                                                                            
  |==============                                                        |  20%
  |                                                                            
  |============================                                          |  40%
  |                                                                            
  |==========================================                            |  60%
               
## Process bandle results
bandleres <- bandleProcess(res)

## Convergence plots
par(mfrow = c(1, 2))
plotConvergence(bandleres)  

#>      [,1] [,2]
#> [1,]    1    2
#> [2,]    2    1