What does it mean (.) in R?

Question

What does it mean (.) in R? As in

sec_axis(~ log(2)/(.),
      breaks = c(2:7,14,21),
      name = "Doubling time (days)")
  )

Besides, it gives an error. The error message is Error: transformation for secondary axes must be monotonic

Thanks!

Answer 1

To give out a proper answer, you should post a reproducible example: data + the ggplot code.

---

If sec_axis() is inside scale_y_continuous, the . is the place holder for the values shown on the main Y axis [if it's in scale_x_continuous, it's the place holder for the values shown on the main X axis.].

---

About the error, I wasn't able to reproduce it. You need to share your ggplot code (a minimal reproducible example that shows the error) and the version you have of ggplot2.

EDIT:

I managed to reproduce the error.

set.seed(1)
df <- data.frame(x = 1:100, y = rnorm(100))

# works!
ggplot(df) +
    geom_line(aes(x = x, y = y))

# doesn't work
ggplot(df) +
    geom_line(aes(x = x, y = y)) +
    scale_y_continuous(sec.axis = sec_axis(trans = ~1/.))
#> Errore: transformation for secondary axes must be monotonic

That happens because zero belongs to the Y axis. In R if you try to divide by zero, it will return Inf, which can't be plotted in an axis. Therefore, the error.

The only way to avoid it is a trick: change the labels and the breaks, without actually applying any transformation. As the example below (obviously this example is meaningless):

ggplot(df) +
    geom_line(aes(x = x, y = y)) +
    scale_y_continuous(sec.axis = sec_axis(~., label = c(-10:-1, 1:10), breaks = 1/c(-10:-1, 1:10)))

In order to reproduce Professor Hyndman's chart, you should do as follow:

library(tidyverse)
library(tsibble)
library(tidycovid19) #remotes::install_github("joachim-gassen/tidycovid19")

updates <- download_merged_data(cached = TRUE)

updates %>%
    mutate(
        cases_logratio = difference(log(confirmed))
    ) %>%
    filter(iso3c %in% countries) %>%
    filter(date >= as.Date("2020-03-01")) %>%
    ggplot(aes(x = date, y = cases_logratio, col = country)) +
    geom_hline(yintercept = log(2)/c(2:7,14,21), col='grey') +
    geom_smooth(method = "loess", se = FALSE) +
    scale_y_continuous(
        "Daily increase in cumulative cases",
        breaks = log(1+seq(0,60,by=10)/100),
        labels = paste0(seq(0,60,by=10),"%"),
        minor_breaks=NULL,
        sec.axis = sec_axis(~ .,
                            labels = c(2:7,14,21),
                            breaks = log(2)/c(2:7,14,21),
                            name = "Doubling time (days)")
    ) +
    ggthemes::scale_color_colorblind()