Question

I am venturing babysteps in ggplots and need some help with the following problem!
I have a dataframe called dftemp which has the following structure

   Year Month Values
1   1996   Jan      0
2   1996   Feb      0
3   1996   Mar      0
4   1996   Apr      0
5   1996   May      0
6   1996   Jun      0
7   1996   Jul      0
8   1996   Aug      0
9   1996   Sep      0
10  1996   Oct      0
11  1996   Nov      0
12  1996   Dec      0

This goes on like this until:

181 2011   Jan     -3
182 2011   Feb      2
183 2011   Mar      2
184 2011   Apr      6
185 2011   May    -13
186 2011   Jun     -4
187 2011   Jul      0
188 2011   Aug      0
189 2011   Sep      0
190 2011   Oct     -7
191 2011   Nov     -7
192 2011   Dec      0

I tried to make a barchart by:

ggplot(dftemp,aes(x=Month,y=Values,fill=Year))+geom_bar(binwidth=10)

And got something weird and wrong (sadly my reputation is not high enough so I cannot post the image)..The resulting plot had y-values that were squished together very tightly by the bottom of the y-axis..

I don’t understand what I am doing wrong here! I tried putting factor() in front of the aes inputs but to no avail!

Answer 1

Your first challenge is to use the stat=”identity” argument in your geom_bar

It is not clear what you are trying to plot, or how you plan to combine your data, so I’ll offer a few options.

First let’s rebuild your data for use:

library(chron)
tt = seq(chron("1/1/1996"),chron("12/31/2011"),by="months")
dftemp = as.data.frame()
dftemp$datestamps = tt
dftemp$years = years(tt)
dftemp$months = months(tt)
dftemp$values = round(runif(length(tt),-15,15))

Here is a simple bar plot, one bar for each date stamp:

ggplot(dftemp,aes(x=datestamps,y=values))+geom_bar(stat="identity",position="dodge")

One bar per month, stacked

ggplot(dftemp,aes(x=months,y=values))+geom_bar(stat="identity")

One bar per year, stacked

ggplot(dftemp,aes(x=years,y=values))+geom_bar(stat="identity")