Create a list of PSV linear models

This function is used to create a list of linear models using the lm() function with phylogenetic species variability (PSV) as the response variable. Requires the function stndrd() from the 'caribmacro' package

psv_LM(
  PSV,
  data,
  x_vars,
  geo_group,
  area = NULL,
  standardize = TRUE,
  log_area = FALSE,
  sq_area = FALSE,
  complete.case = FALSE
)

Arguments

PSV

A data frame in the format that is returned in the first slot of the result of PSV_geo() that holds the PSV values for each species group and geographic feature of interest.

data

A data frame that holds the explanatory variables for each geographic feature of interest.

x_vars

A character or vector equal to the column name(s) in 'data' that hold the explanatory variables of interest.

geo_group

A character equal to the column name in 'SR' and 'data' in which the geographic feature names are stored. This needs to be equal for both the 'SR' and 'data' data frames.

area

Optional. A character or vector equal to the column name in 'data' that holds the area of each geographic feature. This argument is required if the log_area = TRUE and/or sq_area = TRUE.

standardize

Logical. If TRUE then the explanatory variables are standardized by mean centering and dividing by the standard deviation.

log_area

Logical. If TRUE then the geographic area is log transformed by log(area)

sq_area

Logical. If TRUE then the geographic area is squared (i.e. area^2). NOTE: if log_area = TRUE then this option will square log(area).

complete.case

Logical. If TRUE only complete cases will be included in the models. This argument should be set to TRUE if models will be passed to AIC_analysis(). Also, if TRUE then the data used for each model will be exported in the list for use in AIC_analysis().

Examples

if (FALSE) {

library(here)
library(reshape)
library(phyr)

dat <- read.csv(file.path(here(), 'data_raw', 'IBT_Herp_Records_final.csv'), header=TRUE)
geo <- read.csv(file.path(here(), 'data_raw', 'bank_data.csv'), header=TRUE)

coms <- com_matrix(species = "binomial",
                   geo_group = "bank",
                   taxa_group = "class",
                   status = "bnk_status",
                   stat_levels = c("N", "E"),
                   total = TRUE,
                   data = dat)

PSV <- PSV_geo(data = coms, geo_group = "bank")

## Requires the function stndrd()
lm.psv <- PSV_lm(PSV = PSV[['PSV']],
                 data = geo,
                 x_vars = c('area', 'isoPC1', 'isoPC2'),
                 geo_group = 'bank',
                 area = 'area',
                 log_area = TRUE,
                 complete.case = TRUE)

}