[R] ggplot2 stacked line plot

Liam Blanckenberg liam.blanckenberg at gmail.com
Thu Feb 11 03:06:35 CET 2010 

Many thanks for you help on this Dennis.

Liam



On Thu, Feb 11, 2010 at 6:02 AM, Dennis Murphy <djmuser at gmail.com> wrote:
> Hi:
>
> This is closer, but it is still not what you want; it does, however, show
> what
> geom_area is doing when it renders the plot, as can be seen by the colors.
>
> data.set <- data.frame(
>   Time = c(rep(1, 4),rep(2, 4), rep(3, 4), rep(4, 4)),
>   Type = rep(c('a', 'b', 'c', 'd'), 4),
>   Value = c(10, 12, 14, 16, 14, 14, 20, 18, 18, 16, 22, 20, 26, 20, 24, 26)
>  )
> ggplot(data.set, aes(x = Time, y = Value, group = Type)) +
>   geom_area(aes(color = Type), fill = 'transparent', position = 'stack')
>
> You can see that a polygon is drawn (and normally filled) for each group,
> but the line colors are not what you want and the polygon edges are not
> what you want, either. I'd suggest using cumsum() to get cumulative totals
> and using geom_line() instead; it would probably be less work...as in:
>
> library(ggplot2)
> data2 <- ddply(data.set, .(Time), transform, Csum = cumsum(Value))
> ggplot(data2, aes(x = Time, y = Csum, color = Type)) + geom_line()
>
> HTH,
> Dennis
>
> On Tue, Feb 9, 2010 at 8:43 PM, Liam Blanckenberg
> <liam.blanckenberg at gmail.com> wrote:
>>
>> Hi Hadley,
>>
>> Thank you for taking the time to help me with this - I've constructed
>> the following example to illustrate my problem:
>>
>> require(ggplot2)
>>
>> data.set <- data.frame(
>>  Time = c(rep(1, 4),rep(2, 4), rep(3, 4), rep(4, 4)),
>>  Type = rep(c('a', 'b', 'c', 'd'), 4),
>>  Value = c(10, 12, 14, 16, 14, 14, 20, 18, 18, 16, 22, 20, 26, 20, 24, 26)
>> )
>>
>> When I use the following code:
>>
>>   ggplot(data.set, aes(x = Time, y = Value, colour = Type)) +
>> geom_area(aes(fill = Type), position = 'stack')
>>
>> I get a very nice stacked area chart - each layer on the chart
>> represents a Type, with the total height of all the layers
>> representing the total Value across all Types for a given Time period
>> - this is what you would get numerically with the code:
>>
>>   with(data.set, aggregate(Value, by = list(Time), sum))
>>
>> I want to achieve exactly the same thing as this, but instead of
>> having a coloured area to represent each layer I just want a coloured
>> line for each layer. When I use the following code to try and achieve
>> this:
>>
>>   ggplot(data.set, aes(x = Time, y = Value, colour = Type)) +
>> geom_line(aes(colour = Type), position = 'stack')
>>
>> I get a warning/error message: "Missing ymax in position = 'stack'.
>> Maybe you want position = 'identity'?". The plot still renders, but
>> the lines are not stacked - the plot looks exactly as it would if the
>> following code was used instead:
>>
>>   ggplot(data.set, aes(x = Time, y = Value, colour = Type)) +
>> geom_line(aes(colour = Type), position = 'identity')
>>
>> I suspect that the "position = 'stack'" argument is being ignored and
>> 'identity' is being used by default...
>>
>> Hope this helps,
>>
>> Liam
>>
>>
>>
>>
>> On Wed, Feb 10, 2010 at 10:12 AM, hadley wickham <h.wickham at gmail.com>
>> wrote:
>> > Hi Liam,
>> >
>> > Yes, that's what that code should do.  Could you please send a small
>> > reproducible example?
>> >
>> > Hadley
>> >
>> > On Tue, Feb 9, 2010 at 4:21 PM, Liam Blanckenberg
>> > <liam.blanckenberg at gmail.com> wrote:
>> >> Hadley,
>> >>
>> >> Thanks for the pointing that error out to me. Unfortunately, revising
>> >> my code to:
>> >>
>> >> p <- ggplot2(~, aes(x = ~, y = ~, colour = Type)) +
>> >> geom_line(aes(colour = Type), position = 'stack')
>> >>
>> >> still does not generate what I'm after. I'm essentially after a line
>> >> plot where each 'Type' ('series' in excel lingo) is stacked in an
>> >> analogous fashion to a stacked area chart (i.e. geom_area(aes(Fill =
>> >> Type), position = 'stack')) - again using lines rather than areas
>> >> though.
>> >>
>> >> This chart should look just like a stacked area chart, however rather
>> >> than each 'layer' (representing a given 'Type') being filled with a
>> >> colour, each layer should have just a coloured line representing its
>> >> values. Summing each line vertically for a given value on the x-axis
>> >> across all the layers (i.e. 'Types') should then give the total y
>> >> value for that x value (just like vertically summing an area chart).
>> >> Really all I'm after is a stacked area chart where the fill for each
>> >> 'Type' has been removed, leaving only a line for each (stacked) Type's
>> >> value...
>> >>
>> >> I hope this is somewhat clear!
>> >>
>> >> Many thanks,
>> >>
>> >> Liam
>> >>
>> >>
>> >>
>> >>
>> >>
>> >> On Wed, Feb 10, 2010 at 1:53 AM, hadley wickham <h.wickham at gmail.com>
>> >> wrote:
>> >>> Hi Liam,
>> >>>
>> >>> Your syntax is a little off.  You want:
>> >>>
>> >>> p <- ggplot2(~, aes(x = ~, y = ~, colour = Type)) +
>> >>>  geom_area(aes(fill = Type), position = 'stack')
>> >>>
>> >>> Position isn't an aesthetic.
>> >>>
>> >>> Hadley
>> >>>
>> >>> On Sun, Feb 7, 2010 at 10:40 PM, Liam Blanckenberg
>> >>> <liam.blanckenberg at gmail.com> wrote:
>> >>>> Hi all,
>> >>>>
>> >>>> I have been hunting around for hours trying to figure out how to
>> >>>> generate a stacked line chart using ggplot2. This type of chart can
>> >>>> be
>> >>>> generated in excel 2007 by selecting: Chart type > Line > Stacked
>> >>>> line. I can generate a stacked area chart using the following code:
>> >>>>
>> >>>>   p <- ggplot2(~, aes(x = ~, y = ~, colour = Type)) +
>> >>>> geom_area(aes(position = 'stack', fill = Type))
>> >>>>
>> >>>> However, when I try and replicate this using the following code for
>> >>>> geom_line:
>> >>>>
>> >>>>   p <- ggplot(~, aes(x = ~, y = ~, colour = Type)) +
>> >>>> geom_line(aes(position = 'stack'))
>> >>>>
>> >>>> the resulting plot is not stacked - i.e. each 'Type' is plotted at
>> >>>> its
>> >>>> actual value rather than cumulatively to form a stacked chart... I
>> >>>> have poured through Hadley's ggplot2 book (ggplot2: elegant graphics
>> >>>> for data analysis), the R help list and also done general google
>> >>>> searching but cannot find a way to generate this type of plot.
>> >>>>
>> >>>> R version: 2.9.2
>> >>>> ggplot2 version: 0.8.5
>> >>>> OS: windows 7 (64-bit).
>> >>>>
>> >>>> Any suggestions or assistance would be greatly appreciated.
>> >>>>
>> >>>> Regards,
>> >>>>
>> >>>> Liam
>> >>>>
>> >>>> ______________________________________________
>> >>>> R-help at r-project.org mailing list
>> >>>> https://stat.ethz.ch/mailman/listinfo/r-help
>> >>>> PLEASE do read the posting guide
>> >>>> http://www.R-project.org/posting-guide.html
>> >>>> and provide commented, minimal, self-contained, reproducible code.
>> >>>>
>> >>>
>> >>>
>> >>>
>> >>> --
>> >>> http://had.co.nz/
>> >>>
>> >>
>> >
>> >
>> >
>> > --
>> > http://had.co.nz/
>> >
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>
>

More information about the R-help mailing list