CODEWITHSUNDEEP
rggplot2
Marco
Marco

Reputation: 2807

Add regression line via geom_smooth with and without colour grouping in one graph

I would like to illustrate the difference in regression between pooling data (ignoring groups) and controlling for groups. Data comes from http://www.jblumenstock.com/files/courses/econ174/FEModels.pdf

paneldata <- data.frame(Location = c("Chicago", "Chicago", "Peoria", "Peoria", "Milwaukee", "Milwaukee", "Madison", "Madison"),
                           Year = rep(2003:2004, 4), 
                           Price = c(75, 85, 50, 48, 60, 65, 55, 60),
                           Quantity = c(2.0, 1.8, 1.0, 1.1, 1.5, 1.4, 0.8, 0.7))

Since geom_smooth automatically picks the aes I can either do a line for all data points or for groups. But I would like to have both in the same diagram.

library(ggplot2)
library(gridExtra)
plot_pool <- ggplot(paneldata, aes(x=Price, y=Quantity)) + 
  geom_point(aes(colour = Location)) + 
  labs(title="Relationship between Price and Quantity",
       x="Price", y="Quantity") +
  geom_smooth(method = "lm", se = FALSE)  
  
plot_groups <- ggplot(paneldata, aes(x=Price, y=Quantity, colour = Location)) + 
  geom_point() + 
  labs(title="Relationship between Price and Quantity",
       x="Price", y="Quantity") +
  geom_smooth(method = "lm", se = FALSE)  

grid.arrange(plot_pool, plot_groups, ncol=2)

enter image description here

Upvotes: 0

Views: 626

Answers (1)

ek-g
ek-g

Reputation: 721

You just need to add another geom_smooth() with different aesthetics:

ggplot(paneldata, aes(x=Price, y=Quantity)) + 
  geom_point() + 
  labs(title="Relationship between Price and Quantity",
       x="Price", y="Quantity") +
  geom_smooth(aes(colour = Location), method = "lm", se = FALSE) +
  geom_smooth(method = "lm", se = FALSE)

Upvotes: 3

Related Questions