Pairwise Effective Sample Size

Source: R/PairwiseESS.R

This function calculates Effective Sample Size (ESS) on results. The ESS algorithm performs best when results are from multiple runs.

pairwiseESS(inputData)

Arguments

inputData

The input dataset can be a single data frame or a list of data frames.

Value

Returns a matrix with ESS values of all pairwise runs.

Author

Brian O'Meara and Barb Banbury

Examples


data(simRunExample)

# this will give a warning
pairwiseESS(resultsBMExample[[1]]$particleDataFrame)

#> Warning: ESS on a single run should be high; consider combining several runs
#>                  ESS
#> starting_1  55.85303
#> intrinsic_1 88.49379

# ESS should be calculated over multiple runs
pairwiseESS(resultsBMExample)

#> Multiple runs of doRun_prc found, extracting particleDataFrame objects
#>             Runs_1_2
#> starting_1    69.000
#> intrinsic_1  128.526

# you can also manually assemble a list of particleDataFrame tables
    # and use this as the input
inputList <- list(
    resultsBMExample[[2]]$particleDataFrame, 
    resultsBMExample[[1]]$particleDataFrame
    )
pairwiseESS(inputList)

#>             Runs_1_2
#> starting_1   69.0000
#> intrinsic_1 103.7477