Plotting Time Series Values using Bar Chart

Question

I'd like to plot some time series data using a bar chart to show the changes over time, but I'm having trouble doing it with ggplot2.

Sample Data:

df <- structure(list(year = c(1910, 1911, 1912, 1913, 1914, 1915, 1916, 
1917, 1918, 1919, 1920, 1921, 1922, 1923, 1924, 1925, 1926, 1927, 
1928, 1929, 1930, 1931, 1932, 1933, 1934, 1935, 1936, 1937, 1938, 
1939, 1940, 1941, 1942, 1943, 1944, 1945, 1946, 1947, 1948, 1949, 
1950, 1910, 1911, 1912, 1913, 1914, 1915, 1916, 1917, 1918), 
    dday_30 = c(48.3156566976744, 28.4325095348837, 10.9345420465116, 
    30.3018175813953, 22.3314346511628, 6.33566069767442, 27.4823442325581, 
    23.7263461860465, 20.3016930232558, 38.8897564651163, 6.69049065116279, 
    18.4395965581395, 25.8079846046512, 3.57936218604651, 16.6990909767442, 
    21.2175313023256, 15.2260691627907, 9.26520976744186, 9.79716334883721, 
    18.9872194883721, 22.5747300465116, 47.8516541860465, 30.8272055348837, 
    28.0607320930233, 50.1987654883721, 29.4361414418605, 58.4356670232558, 
    45.2513310697674, 29.4801361860465, 44.2652285116279, 38.7353984186047, 
    17.164524372093, 18.8734693488372, 38.7099044186047, 26.9970333953488, 
    13.3448749767442, 31.4876399534884, 29.700936, 26.0668504651163, 
    24.7630490232558, 15.4445029767442, 37.5025793548387, 22.0377372580645, 
    8.63075717741935, 24.8010268548387, 15.4514439516129, 5.60195516129032, 
    21.3893638709677, 18.4909673387097, 19.3719947580645)), .Names = c("year", 
"dday_30"), row.names = c(NA, 50L), class = "data.frame")

Here's what I've tried:

ggplot(stack, aes(x = year, y = dday_30)) + geom_bar()

But's telling me there cannot be a y aesthetic:

Error: stat_count() must not be used with a y aesthetic.

Is there a way to do this with time series data?

?geom_bar: "If you want the heights of the bars to represent values in the data, use stat="identity" and map a variable to the y aesthetic." — Henrik
Your dataset has more than one value for years 1910-1918, how would you like to aggregate those values? Do you want to add them together, average them, or something else? — JasonAizkalns

JasonAizkalns JasonAizkalns · Accepted Answer · 2016-01-28T20:18:09

First, you will need use aggregate to average the dday_30 column by year:

library(ggplot2)

df_melted <- aggregate(data = df, dday_30 ~ year, mean)
names(df_melted) <- c("year", "mean_dday_30")

Then, as pointed out by @Henrik in the comments, you can use stat = "identity" to make the bar graph:

ggplot(df_melted, aes(x = year, y = mean_dday_30)) +
  geom_bar(stat = "identity")

However, for this type of data, a line graph is probably a better choice:

ggplot(df_melted, aes(x = year, y = mean_dday_30)) +
  geom_line() +
  geom_point()

Plotting Time Series Values using Bar Chart

1 Answers