uniglm: Univariate glm regression

In brooksandrew/Rsenal: Rsenal (arsenal) of assorted R functions developed over the years

Description

Runs a univariate logistic regression on each predictor variable of interest.

Usage

uniglm(df, yv, xv, file = NULL, sortby = "aic", xvlab = NULL, test = T)

Arguments

df	data.frame with variables for analysis
yv	character string, target variable
xv	character vector, predictor variables to test univariately
file	character string, filepath to write results out to. txt or csv file.
sortby	character string, criteria to sort variables by. Default = 'aic'
xvlab	data.frame, lookup table for variable names. First column is codename, second column is the pretty printed name.
test	TRUE or FALSE. Includes Chi square test, or not.

Value

data.frame of results

See Also

bestglm, leaps

Examples

require('Hmisc')

##setting up some data
longnames <- data.frame(long = c('Weight', 'Miles Per Gallon', 'Cylinder', 'D.R.A.T', 'Horsepower', 'A.M.'),
short = c('wt', 'mpg', 'cyl', 'drat', 'hp', 'am'), stringsAsFactors=F)

glm.out <- uniglm(df=mtcars, yv='vs', xv=c('hp','drat','cyl','mpg','wt'), xvlab=longnames)

brooksandrew/Rsenal documentation built on May 13, 2019, 7:50 a.m.