Plotting multiple regression lines based on a variable in R ggplot2 and also using a covariate?

I created some sample data ...

example.label <- c("A","A","A","A","A","B","B","B","B","B")
example.value <- c(5, 4, 4, 5, 3, 8, 9, 11, 10, 9)
example.age <- c(30, 40, 50, 60, 70, 30, 40, 50, 60, 70)
example.score <- c(90,95,89,91,85,83,88,94,83,90)
example.data <- data.frame(example.label, example.value,example.age,example.score)

      

What looks like this ...

    example.label example.value example.age example.score
1              A             5          30            90
2              A             4          40            95
3              A             4          50            89
4              A             5          60            91
5              A             3          70            85
6              B             8          30            83
7              B             9          40            88
8              B            11          50            94
9              B            10          60            83
10             B             9          70            90

      

I can plot separate regression lines for "example.value" based on "example.label" by doing this ...

ggplot(example.data, aes(x=example.age,y=example.value,color=example.label)) +
  geom_point() +
  geom_smooth(method = lm,fill=NA)

      

Which looks like this: Graph

However, this just creates a simple linear regression using the 'method = lm' method. The plotting method that I really want for each regression line is calculated using the formula

lm(example.data$example.value ~ example.data$example.age + example.data$example.score)

      

where "example.score" is a covariate. Inside geom_smooth in ggplot can this formula be used to calculate each row? If so, how would I go about it so that it defines corrected linear regressions with covariance?

EDIT: Also, to clarify this, you are asking how to change the formula that geom_smooth in ggplot2 uses to calculate the linear regression from y~x

to y~x+(covariate)

. I am trying to change the way linear regression is plotted, not how to display its equation on a graph.

I hope this was not asked again. Thank.

0


source to share





All Articles