Is there a way to visualize a mutliple regression (more than 3 independent variables)

I am trying to (i) visualise a multiple linear regression (>3 independent variables) and (ii) plot a standard deviation corridor for my linear regression line, but I am not sure how to do this. Specifically, I would like to plot the line - SD. An example with iris:

lm1 <- lm(iris$Sepal.Length ~ iris$Sepal.Width   iris$Petal.Width   iris$Species  iris$Petal.Width)
summary(lm1)
library(car)
avPlots(lm1, intercept = TRUE)
 
library(ggplot2)

ggplot(iris, aes(x = iris$Sepal.Width, y = Sepal.Length,
                   
                   col = factor(Species)))  geom_point(size=1)  theme_bw() geom_smooth(method=lm,se=FALSE, fullrange=TRUE)

Which gives me the added variable plots and the regression line for Sepal.Length~Sepal.Width Species. I was wondering if (i) there is a more effective way than the added variable plots to summarise the results (I have already seen threads like

Example of intended output on the regression line Sepal.Length~Sepal.Width in red:

CodePudding user response：

Maybe something like this? Here, the band represents the standard error, which is a measure of how wrong the regression line is likely to be based on sampling error alone.

library(tidyverse); library(ggforce)
ggplot(iris, aes(x = .panel_x, y = .panel_y))  
  geom_point()  
  geom_smooth(method = "lm")  
  facet_matrix(vars(Sepal.Length:Petal.Width), layer.diag = 3)