ggplot.score_profile: Plot a 'score_profile' object.

In leoguelman/uplift2: Uplift Modeling

Plot a score_profile object.

Description

ggplot method for class ‘score_profile’

Usage

## S3 method for class 'score_profile'
ggplot(x, i.var = NULL, n.type = "boxplot",
  f.type = "percent", statistic = "mean", geom = "point",
  legend.position = NULL, col = NULL, size = NULL, nrows = NULL,
  ncols = NULL, xlab = NULL, ...)

Arguments

x	An object of class score_profile.
i.var	A numeric vector of indices of the variables to plot. The variables should be indexed in the same order that they appear in the initial inspect_balance formula. The default is to plot all variables.
n.type	The type of plot for numeric variables. Boxplots are generated by default. For alternative summary statistics, use summary.
f.type	The type of plot for categorical variables. 100 percent stacked columns are generated by default. The alternative option is counts, which show stacked columns of counts.
statistic	Functions that operate on a vector and produce a single value, as mean and sd do. It may be a user-defined function.
geom	The geometric object to display the data. Argument is passed to ggplot2::stat_summary when type = summary. For type = summary, geom = "point".
legend.position	The position of legends ("left", "right", "bottom", "top"). The default is "right".
col	Color of geom. The default is "red".
size	Size of geom. The default is 2.
nrows	Number of rows for plots.
ncols	Number of columns for plots.
xlab	Title for the x.
...	Additional arguments passed to ggplot2::stat_summary.

Author(s)

Leo Guelman [e-mail]

Examples

set.seed(123)
N <- 10000
eps <- rnorm(N)
age <- round(rnorm(N, 50, 10))
income <- rnorm(N, 60000, 10000) + 200 * age
gender <- gl(2, N/2,  labels = c("F", "M"))
insurance <- gl(4, N/4,  labels = c("HOME", "AUTO", "LIFE", "HEALTH"))
z <- 1e-01 + 0.1 * age - 1e-04 * income + 0.3 * (gender == "F") + eps
pr <- 1 / (1 + exp( -z))
purchase <- rbinom(N, 1, pr)
df <- data.frame(purchase, age, income, gender, insurance)
### Fit glm
pred <- fitted(glm(purchase ~ age + income + gender,
                  data = df, family = "binomial"))
profileForm <- pred ~ age + income + gender + insurance
prof1 <- score_profile(profileForm, data = df)
prof1
ggplot(prof1)

leoguelman/uplift2 documentation built on April 15, 2022, 4:34 a.m.