ezlm

The goal of ezlm is to fit a simple linear regression model, outputting easy-to-read results rounded to user’s preferred decimal place. A wrapper for lm().

Installation

The ezlm package is not yet available from CRAN. Please download it from this repository using the following R command:

devtools::install_github("deansn/ezlm")

Example

Fuel Economy and Horsepower inmtcars

As a basic demonstration of the ezlm() function, begin by calling the function and inputting the appropriate variables (as described in the documentation above).

Let’s say we’re interested in the effect of fuel economy on horsepower, and we want to see our results to three decimal places:

library(ezlm)
ezlm(mtcars, mtcars$mpg, mtcars$hp, 3)
#> [1] "Regression complete! The intercept of your model is 324.082, with a p-value of 0. The coefficient for your independent variable is 27.433, with a p-value of 0. Overall, your model has an r-squared value of 0.602."

Urban Population and Assaults in USArrests

While this function was designed primarily for linear regressions within the mtcars dataset, it should work with other datasets, such as USArrests. In this example, let’s look at the influence of the urban population proportion on the number of assault charges.

ezlm(USArrests, USArrests$UrbanPop, USArrests$Assault, 2)
#> [1] "Regression complete! The intercept of your model is 73.08, with a p-value of 0.18. The coefficient for your independent variable is 53.85, with a p-value of 0.07. Overall, your model has an r-squared value of 0.07."

Inappropriate Inputs

To force my function to throw one of my custom error messages, we can try to replace one of the numerical parameters with a non-accepted input class:

ezlm(mtcars, as.factor(mtcars$mpg), mtcars$hp, 3)
#> Error in ezlm(mtcars, as.factor(mtcars$mpg), mtcars$hp, 3): Sorry! Your IV must be numeric.
#>          It's currently of class factor

As expected, the function throws the custom error message when the IV is defined as a factor variable.