contact (version 1.2.8)

randomizeFeature: Randomize or Pseudorandomize Categorical Variables

Description

This function randomizes the values in a given column (or set of columns (i.e., c(colname(x)[1], colname(x)[2]))), identified by the "feature" argument in a dataset (x).

Usage

randomizeFeature(
  x,
  feature = NULL,
  shuffle = FALSE,
  maintainDistr = TRUE,
  numRandomizations = 1
)

Arguments

x

Data frame containing real-time-location data.

feature

Vector of 1 or more column names describing variables in x to be randomized.

shuffle

Logical. If TRUE, unique values will be replaced with another, random unique value from the same distribution with 100 certainty. For example if the values in a "dose" column c(0mg,100mg,300mg) were shuffled, one possible outcome would be: x$dose.shuff[which(x$dose == "0mg")] <- 300mg, x$dose.shuff[which(x$dose == "100mg")] <- 0mg, and x$dose.shuff[which(x$dose == "300mg")] <- 100mg. Defaults to FALSE.

maintainDistr

Logical. If TRUE, the number of each unique value in the column will be maintained in the function output. Otherwise, the function will draw on the initial distribution to assign randomized values, but the specific number of each unique value may not be maintained. Defaults to TRUE.

numRandomizations

Integer. The number of replicate data frames produced in output. Defaults to 1.

Value

Output is x appended with columns described below.

...shuff

Randomized value of specified variables.

randomRep

Randomization replicate.

Details

Note: the shuffle argument supercedes the maintainDistr argument. Therefore, if shuffle == TRUE, the maintainDistr argument is irrelevant.

References

Farine, D.R., 2017. A guide to null models for animal social network analysis. Methods in Ecology and Evolution 8:1309-1320. https://doi.org/10.1111/2041-210X.12772.

Examples

# NOT RUN {
data(calves)

system.time(randomizedValues<-contact::randomizeFeature(x = calves, 
   feature = c("calftag", "date"), shuffle = TRUE, maintainDistr = TRUE, 
   numRandomizations = 3)) 
   
randomizedFrame<-data.frame(randomizedValues[[1]], stringsAsFactors = TRUE)

head(randomizedFrame) #see that randomized-value columns have been appended.
# }