Filter a set of correlated predictors to fewer ones — predictor

This function helps to remove highly correlated variables from a set of predictors. It supports multiple options some of which require both environmental predictors and observations, others only predictors.

Some of the options require different packages to be pre-installed, such as ranger or Boruta.

Usage

predictor_filter(env, keep = NULL, method = "pearson", ...)

Arguments

env: A data.frame or matrix with extracted environmental covariates for a given species.
keep: A vector with variables to keep regardless. These are usually variables for which prior information is known.
method: Which method to use for constructing the correlation matrix (Options: 'pearson' (Default), 'spearman'| 'kendal'), "abess", or "boruta".
...: Other options for a specific method

Value

A character vector of variable names to be excluded. If the function fails due to some reason return NULL.

Details

Available options are:

"none" No prior variable removal is performed (Default).
"pearson", "spearman" or "kendall" Makes use of pairwise comparisons to identify and remove highly collinear predictors (Pearson's r >= 0.7).
"abess" A-priori adaptive best subset selection of covariates via the abess package (see References). Note that this effectively fits a separate generalized linear model to reduce the number of covariates.
"boruta" Uses the Boruta package to identify non-informative features.

Note

Using this function on predictors effectively means that a separate model is fitted on the data with all the assumptions that come with in (e.g. linearity, appropriateness of response, normality, etc).

Examples

if (FALSE) { # \dontrun{
 # Remove highly correlated predictors
 env <- predictor_filter( env, option = "pearson")
} # }