Question

I am creating a times series using ggplot2 in R. I would like to know how to show tick marks in the x-axis only for the months that are labeled (e.g. Mar 07, Mar 08, etc) while keeping the vertical grey lines for every single month.

The main reason is because having a tick mark for every month makes it hard to know which one correspond to the labels.

Here is an example of a plot:

Here is the line of R behind:

ggplot(timeseries_plot_data_mean,aes(as.numeric(project_date)))+
   geom_line(aes(y=num_views))+geom_point(aes(y=num_views))+
   stat_smooth(aes(y=num_views),method="lm")+
   scale_x_continuous(breaks = xscale$breaks, labels = xscale$labels)+
   opts(title="Monthly average num views")+xlab("months")+ylab("num views")

This is what would like to generate. See how the ticks are positioned right above the month label and the vertical lines are still there showing each month.

I manually edited the plot above using Inkscape, (ignore the q’s, Inkscape strangely replaced the dots for q’s)

Answer 1

Here is a solution using the minor_breaks parameter of scale_x_date(). To use this, your x-values must be of class Date instead of numeric.

library(ggplot2)
set.seed(123)

x <- seq(as.Date("2007/3/1"), as.Date("2012/4/1"), by = "1 month")
y <- ((exp(-10 * seq(from=0, to=1, length.out=length(x))) * 120) +
      runif(length(x), min=-10, max=10))

dat <- data.frame(Months=x, Views=y)

x_breaks <- seq(as.Date("2007/3/1"), as.Date("2012/4/1"), by="1 year")
x_labels <- as.character(x_breaks, format="%h-%y")

plot_1 <- ggplot(dat, aes(x=Months, y=Views)) +
          theme_bw() +
          geom_line() +
          geom_point() +
          scale_x_date(breaks=x_breaks, labels=x_labels, minor_breaks=dat$Months)

png("plot_1.png", width=600, height=240)
print(plot_1)
dev.off()