R: Setting Tick Label Limits on Faceted Plot with Free Y-Axis

Question

I'm relatively new to R and ggplot, but I've already found a common issue occurring with my plots

For some reason, ggplot usually decides not to show the top tick label on the y-axis. I found the below image from a random Google search, but it illustrates the problem that I'm referring to:

As you can see from the above chart, one of the lines extends well above the top tick label on the y-axis. I find this very frustrating and it seems strange that to me that ggplot does this. I would much prefer all of the data in my charts to be contained within the minimum and maximum tick labels, rather than stretching beyond.

Usually this wouldn't be a big issue, since one can just manually configure the y-axis labels. However in this instance I have two facets with the scale on the y-axes set the free_y. Consequently, I can't just set the y-axis tick labels manually, since they need to have differing values for either plot. Here are the plots:

Ideally, on the left plot, the y-axis' top tick label would be 40, and on the right plot it would be 6. This would ensure that all of the data are contained within the y-axis' minimum and maximum limits, but due to the aforementioned complication introduced by faceting, this seems to be easier said than done.

Is it possible for me to achieve the desired effect without completely revamping the code?

Here is a link to the CSV files that I'm using: https://onedrive.live.com/?authkey=%21AEeTM7phVBNGI5c&id=ACB3DC15E10D8AF1%213433&cid=ACB3DC15E10D8AF1. Unfortunately, when sharing CSVs via OneDrive, they get converted into excel files. So you may have to access these files and export them into CSVs in order for the following code to work:

library(ggplot2)
library(scales)
library(extrafont)
library(dplyr)
library(tidyr)

# you may also need to adjust the working directory here:
work_dir <- "D:\\OneDrive\\Documents\\Economic Data\\Historical Yields\\Eurozone"
setwd(work_dir)

germany_yields <- read.csv(file = "Germany 10-Year Yield Weekly (2007-2020).csv", stringsAsFactors = F)
germany_yields <- germany_yields[, -(3:6)]
colnames(germany_yields)[1] <- "Date"
colnames(germany_yields)[2] <- "Germany.Yield"

italy_yields <- read.csv(file = "Italy 10-Year Yield Weekly (2007-2020).csv", stringsAsFactors = F)
italy_yields <- italy_yields[, -(3:6)]
colnames(italy_yields)[1] <- "Date"
colnames(italy_yields)[2] <- "Italy.Yield"

greece_yields <- read.csv(file = "Greece 10-Year Yield Weekly (2007-2020).csv", stringsAsFactors = F)
greece_yields <- greece_yields[, -(3:6)]
colnames(greece_yields)[1] <- "Date"
colnames(greece_yields)[2] <- "Greece.Yield"

combined <- merge(merge(germany_yields, italy_yields, by = "Date", sort = F), 
                  greece_yields, by = "Date", sort = F)

combined <- na.omit(combined)
combined$Date <- as.Date(combined$Date,format = "%B %d, %Y")
combined["Italy_v_Germany.Spread"] <- combined$Italy.Yield - combined$Germany.Yield
combined["Greece_v_Germany.Spread"] <- combined$Greece.Yield - combined$Germany.Yield

fl_dates <- c(tail(combined$Date, n=1), head(combined$Date, n=1))

longcombined <- gather(combined, 
                       key="measure", 
                       value="value", 
                       c("Italy_v_Germany.Spread",
                         "Greece_v_Germany.Spread"))

ggplot(data=longcombined, aes(x = Date, y = value)) + geom_line() +

       facet_wrap(~measure, scales = "free_y") +

       geom_blank(aes(y = 0)) +

       scale_x_date(limits = fl_dates,
                    breaks = seq(as.Date("2008-01-01"), as.Date("2020-01-01"), by="2 years"),
                    expand = c(0, 0),
                    date_labels = "%Y") +

       scale_y_continuous(n.breaks = 7)

score 2 · Answer 1 · answered Apr 30 '20 at 13:47

Here's a bit of a hacky way of doing it. Before you call ggplot, create a two-row data frame containing one row each from your two faceting levels, and change the values in the two rows to just above the upper threshold you would like set for each:

hackdf <- longcombined[which(!duplicated(longcombined$measure)),]
hackdf$value <- c(6.1, 41)

Now add plot an invisible point on each plot with the line geom_point(data = hackdf, alpha = 0) :

Anurag N. Sharma · Answer 2 · 2020-04-30T15:38:16.777

Have you tried @Uwe's answer at: ggplot2 change axis limits for each individual facet panel? It wouldn't require you to edit existing code, just add the facet_grid_sc layer.

It got the job done for me.

Here is the default behaviour (adapted from @Uwe's answer linked here):

devtools::install_github("zeehio/facetscales")
library(facetscales)
library(ggplot2)

#default behaviour

ggplot(mpg, aes(displ, cty)) + 
  geom_point() + 
  facet_grid(drv~.)

Now when the facet y-axis is to be edited individually, add:

scales_y <- list(
  `4` = scale_y_continuous(limits = c(5, 25), breaks = seq(5, 25, 5)),
  `f` = scale_y_continuous(limits = c(0, 40), breaks = seq(0, 40, 10)),
  `r` = scale_y_continuous(limits = c(10, 20), breaks = seq(10, 20, 2))
)

# y-axis tick labels edited in individual facets

ggplot(mpg, aes(displ, cty)) + 
  geom_point() +
  facet_grid_sc(rows = vars(drv), 
                scales = list(y = scales_y))

You can add the desired tick labels by editing both limits and breaks in scale_y_continuous.

Haven't had much luck with this as I'm getting all sorts of errors and can't find any meaningful explanations of how facet_grid_sc works. Can you point to any resources which help explain how to use this? — Ryan Walter, Apr 30 '20 at 15:10
The easiest thing to do here would be to go through the code step by step and run the intermediate steps to understand what happens. Reading documentation goes a long way as well. And regarding ```vars```, you should refer to Hadley Wickham's dplyr package vignette and examples. ```drv``` is a variable within the ```mpg``` object. Good luck! — Anurag N. Sharma, Apr 30 '20 at 15:55

R: Setting Tick Label Limits on Faceted Plot with Free Y-Axis

2 Answers2