Empirical permutation test of significance of scoreFn(modelValues,yValues) >= scoreFn(modelValues,perm(yValues)).

Source: R/EmpiricalFns.R

Treat permutation re-samples as similar to bootstrap replications.

permutationScoreModel(
  modelValues,
  yValues,
  scoreFn,
  ...,
  na.rm = FALSE,
  returnScores = FALSE,
  nRep = 100,
  parallelCluster = NULL
)

Arguments

modelValues

numeric array of predictions.
yValues

numeric/logical array of outcomes, dependent, or truth values
scoreFn

function with signature scoreFn(modelValues,yValues) returning scalar numeric score.
...

not used, forces later arguments to be bound by name
na.rm

logical, if TRUE remove NA values
returnScores

logical if TRUE return detailed permutedScores
nRep

integer number of repititions to perform
parallelCluster

optional snow-style parallel cluster.

Value

summaries

Examples


set.seed(25325)
y <- 1:5
m <- c(1,1,2,2,2)
cor.test(m,y,alternative='greater')

#> 
#> 	Pearson's product-moment correlation
#> 
#> data:  m and y
#> t = 3, df = 3, p-value = 0.02883
#> alternative hypothesis: true correlation is greater than 0
#> 95 percent confidence interval:
#>  0.1526678 1.0000000
#> sample estimates:
#>       cor 
#> 0.8660254 
#> 
f <- function(modelValues,yValues) cor(modelValues,yValues)
permutationScoreModel(m,y,f)

#> [1] "Studentized permutation test: is observed score greater than permuted score,  summary: p=0.04162"