我正在使用以下数据结构来尝试制作堆积区域图表:
df <- data.frame(PopDen = c( 0.002279892, 0.002885407, 0.004291351, 0.002457731, 0.006631572, 0.007578882, 0.004465446, 0.007436628, 0.009001456, 0.006951703, 0.003602076, 0.005695585, 0.005819783, 0.007412274, 0.004931548, 0.006257411, 0.008635908, 0.005438558, 0.002251421,0.006438558), DomArea = c( 253500, 135270, 197180, 131590, 142210, 166920, 125640, 184600, 139940, 126280, 127760, 190940, 133440, 143510, 117260, 69340, 143620, 127480, 181970,164180), PR_Cat = c( "High", "High", "Low", "Low", "Low", "Low", "Low", "Low", "High", "High", "Medium", "Medium", "Medium", "Low", "Low", "Medium", "Medium", "Low", "Low","Low") )
p <- ggplot(df, aes(PopDen, order(DomArea), colour = PR_Cat))
p + geom_area(aes(colour = PR_Cat, fill= PR_Cat), position = 'stack')
但是,我不明白如何将这些区域叠加在一起;目前他们正在重叠。我假设我在这里需要一个position = 'stack'
参数,但无论是否包含,情节看起来都是一样的。
此外,是否可以按DomArea
中的某个类别订购PR_Cat
,还是需要重新组织我的数据?
答案 0 :(得分:23)
我不确定你在这里绘制什么,但是你不想在 y 轴上绘制PopDen
而不是 x 轴?您可以使用DomArea
包中的PR_Cat
按每个ddply
类别订购plyr
,然后堆叠的工作方式如下:
的 修改 强>
我意识到你可能希望按照Low, Medium High
的顺序堆积图,所以我们需要首先在PR_Cat
因子上强制执行这个排序:
df$PR_Cat <- ordered( df$PR_Cat, levels = c('Low', 'Medium', 'High'))
现在使用DomAreaByCat
创建ddply
列:
df <- ddply(df, .(PR_Cat), transform, DomAreaByCat = order(DomArea))
您的df
将如下所示:
> df
PopDen DomArea PR_Cat DomAreaByCat
1 0.004291351 197180 Low 8
2 0.002457731 131590 Low 5
3 0.006631572 142210 Low 9
4 0.007578882 166920 Low 2
5 0.004465446 125640 Low 3
6 0.007436628 184600 Low 7
7 0.007412274 143510 Low 11
8 0.004931548 117260 Low 4
9 0.005438558 127480 Low 10
10 0.002251421 181970 Low 6
11 0.006438558 164180 Low 1
12 0.003602076 127760 Medium 4
13 0.005695585 190940 Medium 1
14 0.005819783 133440 Medium 3
15 0.006257411 69340 Medium 5
16 0.008635908 143620 Medium 2
17 0.002279892 253500 High 4
18 0.002885407 135270 High 2
19 0.009001456 139940 High 3
20 0.006951703 126280 High 1
然后你可以像这样做堆积区域图:
p <- ggplot(df, aes( DomAreaByCat, PopDen))
p + geom_area(aes(colour = PR_Cat, fill= PR_Cat), position = 'stack')