stat_count for month-year combination

Question

I want to plot the number of occurrences of a certain date in a data frame. I write the following:

dates <-  c("2018-01-03", "2018-01-17", "2018-02-03", "2018-02-03", "2018-02-16", 
            "2018-02-26", "2018-03-03")
data <- as.data.frame(dates) 

ggplot(data) +
  stat_count(aes(x = (dates)))

Basically the code plots the number of occurences of each day-month-year combination in a date. Now, I want to create a plot that counts the number of occurences of each month-year combination in a date. Basically, considering the picture, bars that have the same month-year should be one above the other. Does anyone have any idea?

Answer 1

library(tidyverse)
dates <-  c("2018-01-03", "2018-01-17", "2018-02-03", "2018-02-03", "2018-02-16", 
            "2018-02-26", "2018-03-03")
data <- as.data.frame(dates) 

data <- mutate(data,
               dates = lubridate::as_date(dates),
               year_month=paste(lubridate::year(dates),
                                str_pad(lubridate::month(dates), width = 2, pad = 0),
                                sep="-"))

ggplot(data) +
  stat_count(aes(x = year_month))

If you want the days to be distinguishable (not sure how useful this might be with a real world example) you can use fill , notice factor(dates) so that dates is treated as a factor.

ggplot(data) +
  stat_count(aes(x = year_month, fill=factor(dates)))