README.md

In jesprabens/ggregplt: Creates interactive plots of regression models.

ggregplt

The package, ggregplt, uses ggplot and plotly to make interactive regression plots, diagnostic plots, and correlation matrices for regression.

This package can be downloaded using the following steps:

library(devtools)
install_github("jesprabens/ggregplt")

library(ggregplt)

int_reg - Interactive Regression Plot

This function allows the user to input a regression model and get an itneractive regression plot returned. Moreover, if a simple linear model is plotted this function can also take in an x value to produce a predited point on the plot. Multiple linear regression models can also be plotted, but not with an inputted x.

Example: int_reg

regression <- lm(Sepal.Width~Sepal.Length, data = iris)

int_reg(mod = regression, x = 3)

This will prodcue an interactive plot with a predicted value of Sepal.Width when Sepal.Length is 5.

VIF_table

This function allows the user to plug in a multiple regression equation along with its' data set and gives the corresponding VIF values with the correlation matrix in order to diagnose multicollinearity

Example: VIF_table

equation <- lm(Sepal.Width~Sepal.Length+Petal.Length, data = iris)
VIF_Table(equation, 3)

This example will produce a dataframe of the VIF values of the predictor variables predicting Sepal.Width based on Sepal.Length, and Petal.Length from. the iris dataset and a note telling the user whether or not the VIF values are below their given threshold.

MR - Multiple Regression

This function allows the user to plug in a multiple regression equation and returns a corresponding interactive correlation matrix in order to diagnose which variables are causing multicollinearity.

Example: MR

equation <- lm(Sepal.Width~Sepal.Length+Petal.Length, data = iris)
MR(iris, equation)

This example will produce an interactive correlation matrix of the predictor variables predicting Sepal.Width based on Sepal.Length, and Petal.Length from. the iris dataset.

jesprabens/ggregplt documentation built on May 12, 2020, 4:53 p.m.