How to write both superscript and subscript text using geom_text in R?

Question

I'd like to write a model equation in the graph.

nitrogen<- c(10,15,20,25)
yield<- c(50,70,85,30)
dataA<- data.frame(nitrogen,yield)

ggplot(data=dataA, aes(x=nitrogen, y=yield))+
  geom_point (col="Black", size=4) +
  stat_smooth(method='lm', linetype=1, se=FALSE, formula=y~poly(x,2, raw=TRUE), size=0.5, color="Blue") +
  windows(width=5.5, height=5)

Now I'd like to write below equation inside the graph. I know there are many questions about this, but I cannot find a solution, and still struggling. So I ask again. Sorry for duplicated question.

Could you please let me know how to write both superscript and subscript text?

Always many thanks!!

Where? As an axis label, as a `geom_text`, an an `annotate`ion, a title, or something else? — r2evans, Nov 02 '22 at 22:16
To write math in general, see [`help("plotmath")`](https://stat.ethz.ch/R-manual/R-devel/library/grDevices/html/plotmath.html). To add the regression equation to a `ggplot` graph, see [here](https://stackoverflow.com/questions/7549694/add-regression-line-equation-and-r2-on-graph). I don't hammer close as a duplicate because of the subscripts part of the question. — Rui Barradas, Nov 02 '22 at 22:34

Ric · Accepted Answer · 2022-11-02T23:10:16.697

2

library(ggplot2)

nitrogen<- c(10,15,20,25)
yield<- c(50,70,85,30)
dataA<- data.frame(nitrogen,yield)
coef = coef(lm(yield ~ poly(nitrogen,2, raw=TRUE)))

ggplot(data=dataA, aes(x=nitrogen, y=yield))+
  geom_point (col="Black", size=4) +
  stat_smooth(method='lm', linetype=1, se=FALSE, formula=y~poly(x,2, raw=TRUE), size=0.5, color="Blue") +
  windows(width=5.5, height=5) + 
  annotate("text", x=15, y=16, label = bquote(~ Y[i][j]==.(coef[1])+.(coef[2])*x+.(coef[3])*x^2))

Edit This version gives more control on formula display:

ggplot(data=dataA, aes(x=nitrogen, y=yield))+
  geom_point (col="Black", size=4) +
  stat_smooth(method='lm', linetype=1, se=FALSE, formula=y~poly(x,2, raw=TRUE), size=0.5, color="Blue") +
  windows(width=5.5, height=5) + 
  annotate("text", x=15, y=16, label = bquote(
    ~ paste(Y[i][j],"=", .(sprintf("%.2f", coef[[1]])), .(sprintf("%+.2f",coef[[2]])),x, .(sprintf("%+.4f",coef[[3]])),x^2)))

edited Nov 02 '22 at 23:10

answered Nov 02 '22 at 22:45

Ric

5,362
1
10
23

Thank you so much!! How about writing inside graph? i.e. x=20, y=70 ? – Jin.w.Kim Nov 02 '22 at 22:51
just use `annotate("text", x=15, y=15, label = bquote(~ Y[i][j]==.(coef[[1]])+.(coef[[2]])*x+.(coef[[3]])*x^2))` instead of `ggtitle`, and adjust x and y – Ric Nov 02 '22 at 22:55
If you prefer more control to numeric format (also prevent the last "+-" in the formula), you can use sprintf as follows: `label = bquote(~ paste(Y[i][j],"=", .(coef[[1]]), .(sprintf("%+f",coef[[2]])),x, .(sprintf("%+f",coef[[3]])),x^2))` – Ric Nov 02 '22 at 23:02
If I use annotate(), it says Error:! All columns in a tibble must be vectors. x Column `label` is a call. Could you check this for me? – Jin.w.Kim Nov 02 '22 at 23:02
But thank you so much!! The reason I wanted to write an equation was when I use 'stat_poly_eq()', the equation model does not reflect blocks. So model equation inside graph was different from my model, 'lm(y~poly(x,2,raw=TRUE)+ factor(Rep)'. So I was trying to write directly. this code '(coef[1])+.(coef[2])*x+.(coef[3])*x^2)' is so nice. I can reflect blocks now when I set up like 'coef(lm(y~poly(x,2,raw=TRUE)+ factor(Rep)' Thank you so much!!! You saved my whole day. – Jin.w.Kim Nov 02 '22 at 23:08
I edited the example to put the formula inside the plot. – Ric Nov 02 '22 at 23:14

How to write both superscript and subscript text using geom_text in R?

1 Answers1