I have created one monthly plot with facet_wrap
.
So in the plot I have 3 rows and 4 columns. Now I want to set my common y axis for each rows e.g 1st row should have one common y values, same goes with the 2nd and 3rd rows.
I tried but not able to do it.
I used
ggplot(data = PB,
aes(x = new_date, y = Mean, group = 1)) +
geom_line(aes(color = experiment)) +
theme(legend.title = element_blank()) +
facet_wrap( ~MonthAbb, ncol = 4, scales = "free")
The issue is the scales = "free". Remove this and it will set a common scale across rows and columns (or use "free_y" or "free_x" to adjust accordingly).
If what you're looking for is a separate scale for each row, it will require a bit more work. Check this solution at R: How do I use coord_cartesian on facet_grid with free-ranging axis which layers invisible points on the plot to force the look you want. Otherwise a simple solution might to look at using gridExtra and plot each row separately, then merge into a grid.
Edit: a gridExtra solution would look something like:
library(gridExtra)
g1 <- ggplot(data = PB1, aes(x=new_date, y = Mean, group = 1)) +
geom_line(aes(color = experiment)) +
theme(legend.title = element_blank())
g2 <- ggplot(data = PB2, aes(x=new_date, y = Mean, group = 1)) +
geom_line(aes(color = experiment)) +
theme(legend.title = element_blank())
grid.arrange(g1, g2, nrow=2)
Here is an option to set these on a per-panel basis. It is based on a function I've put in a github package. I'm using some dummy data as example.
library(ggplot2)
library(ggh4x)
df <- data.frame(
x = rep(1:20, 9),
y = c(cumsum(rnorm(60)) + 90,
cumsum(rnorm(60)) - 90,
cumsum(rnorm(60))),
row = rep(LETTERS[1:3], each = 60),
col = rep(LETTERS[1:3], each = 20)
)
ggplot(df, aes(x, y)) +
geom_line() +
facet_wrap(row ~ col, scales = "free_y") +
facetted_pos_scales(
y = rep(list(
scale_y_continuous(limits = c(90, 100)),
scale_y_continuous(limits = c(-100, -80)),
scale_y_continuous(limits = c(0, 20))
), each = 3)
)
