Indicator.xPerm: Permutates and calculates for each permutation both the non...
In Zabados/PhiCor: PhiCor: Calculating the Phi coefficient of Association

Description Usage Arguments Value Author(s) See Also Examples

This function purmutaes and calculates the non and group-equalised versions of species indicator value (IndVal) De Cáceres and Legendre (2009). It is useful as it gives the distribution of IndVal values if the species are distributed randomly across the locations, the IndVal value for the actual data is compared to this distribution to calculate the p-value. The distribution is a useful output to visualise the distribution.

Indicator.xPerm(
  InDataframe,
  FieldToPerm,
  weighted,
  groupsField,
  targetGroup,
  x = 1000,
  SquareID = NULL,
  toroidal = FALSE
)

`InDataframe`	dataframe - containing the data on species presence absence, habitat and weighting. It also optionally could contain location ID's.
`FieldToPerm`	string - the name of the binary field within InDataframe that contains binary presence absence for a species at the location.
`weighted`	string - name of habitat weighting field. For example, proportion of habitat, but any weighting with all habitats at a location adding to 1.
`groupsField`	string - ID of habitats or groups of interest.
`x`	Integer - optional - the number of permutation, set as 1000 by default.
`SquareID`	string - optional if non-weighted, but MUST be used if a weighting is used. ID of each location. Useful later for permutating the locations to calculate p-values.
`toroidal`	Boolean - If you suspect that your data is spatially auto-correlated you should use a toroidal permutation.
`targetgroup`	Integer - ID of the habitat of interest.

Two data frames, one for the group-equalised $'GroupEqualPermList' and one for the non group-equalised $NonGroupEqualPermList. It also returns the number of permutations $Permutations.

Jordan Chetcuti

PhiCor

    #Building on the examples from \link[PhiCor]{Indicator.nonEqual}
    #Using the included dataframe Species1
    Inputs_species1 = CorIndex(InDataframe = Species1, speciesbinary = 'Species1', weighted = 'Proportion', group = 'HabId', 'LocationID')
    Inputs_species1$nkoverNk
    #Using CorIndex.TargetVar, the input is the output of Indicator. In this case we are interested in
    #habitat 1 (using the ID for the habitat)
    Inputs_species1_hab1 = CorIndex.TargetVar(Inputs_species1, 1)

    species1_hab1_Indval = Indicator.nonEqual(Inputs_species1_hab1)

    #histogram showing the output of Indicator.xPerm for habitat 1 in comparison to the IndVal value from actual data.
    hist(Hab1_PermutatedPhi$NonGroupEqualPermList, main = "Non group-equalised distribution", xlab = "IndVal values", breaks = 100)
    abline(v=species1_hab1_IndVal, col="red")
    text((species1_hab1_IndVal + 0.006), 15, "Calulated IndVal value", col = "red", srt=90)