crossplot_stats: Basic Statistics for regression varaibles and plotting
In ChandlerLutz/crossplotr: Easily create cross-sectional plots with ggplot2

Description Usage Arguments Value

View source: R/crossplot_stats.R

This function will plot either an unweighted or weighted regression line and print summary statistics. Log regressions are also permitted

1
2
3

crossplot_stats(p, log.reg = FALSE, weighted = FALSE,
  sprintf.format = "%.2f", xlabel = NULL, ylabel = NULL,
  reg.label.se = TRUE, mean.label.se = FALSE)

`p`	a crossplot `ggplot2` plot. Can be generated from `crossplot`
`log.reg`	logical variable that indicates if the regression should be run in logs. Defaults to `FALSE`
`weighted`	logical variable that indicates if the the regression will be weighted by the `size` variable from the plot. Defaults to `FALSE`
`sprintf.format`	the string format for the stats in `sprintf`.
`xlabel`	a string with the name for the x-variable. Defaults to `NULL`. If set to `NULL`, the variable name from the x-axis will be used
`ylabel`	a string with the name for the y-variable. Defaults to `NULL`. If set to `NULL`, the variable name from the y-axis will be used
`reg.label.se`	logical. If set to `TRUE`, the regression standard errors will be printed
`mean.label.se`	logical. If set to `TRUE`, the standard errors for the means will be printed.

a list of two data.frames with the crossplot stats. The first element of the list, plot.stats holds a data.frame of the stats in numeric format. The second element of the list plot.stats.pretty holds a data.frame of the stats in a pretty stringr format using sprintf. The data.frame will include the following variables: "slope", "slope.se", "intercept", "intercept.se", "r.squared", "adj.r.squared", "mean.x", "mean.y" "var.x", "var.y", "sd.x", "sd.y". The standard errors for the slope and the intercept will be printed in parentheses next to the coefficient in the labels. For the labels of the means, variances, or standard deviations, xlabel and ylabel will be used for the labels. Standard errors will be heteroskedasticity-robust white standard errors