BalancedGrps: Function for resampling groups to equal sample size

In ArdernHB/KnnDist: Kernal Nearest Neighbour classification with distance inputs

Description

This function takes a character of factor vector and returns a new vector of equal sample size along with a vector of index positions to sort whatever other data might be associated with the factor vector into the new equal sample size order. The user can also provide a user defined minimum sample size if the user wishes the resampling procedure to sample to a smaller number of specimens than the smallest sample size. In this way the function can also be used in a leave-p-out cross- validation exercise.

Usage

BalancedGrps(GroupMembership, GroupSize = NA)

Arguments

GroupMembership	a character or factor vector in the same order as the distance data to denote group membership.
GroupSize	the sample size you wish the function to resample the data to. If NA supplied then the smallest sample size of the groups supplied will be used.

Value

Returns a list of 2 vectors. The first vector is the indexed location of the new equally size sample

Author(s)

Ardern Hulme-Beaman

ArdernHB/KnnDist documentation built on Feb. 5, 2021, 5:09 a.m.