拿R來畫畫（八）：面積圖與堆積圖

面積圖

面積圖主要用來展示數量隨時間變化的情況，單一變量的面積圖在數字起伏較大時視覺效果較好。

平平無奇面積圖

library(ggplot2)
library(gcookbook)
# 太陽黑子數據集
sunspots

A Time Series: 235 × 12

	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
1749	58.0	62.6	70.0	55.7	85.0	83.5	94.8	66.3	75.9	75.5	158.6	85.2
1750	73.3	75.9	89.2	88.3	90.0	100.0	85.4	103.0	91.2	65.7	63.3	75.4
1751	70.0	43.5	45.3	56.4	60.7	50.7	66.3	59.8	23.5	23.2	28.5	44.0
1752	35.0	50.0	71.0	59.3	59.7	39.6	78.4	29.3	27.1	46.6	37.6	40.0

# 將timeseries數據轉換成dataframe數據
dss = data.frame(year = as.numeric(time(sunspot.year)), ss = as.numeric(sunspot.year))
head(dss)

A data.frame: 6 × 2

year	ss
<dbl>	<dbl>
1700	5
1701	11
1702	16
1703	23
1704	36
1705	58

ggplot(dss, aes(x = year, y = ss)) + 
geom_area()

精緻優雅面積圖

# color指定邊框顏色，fill指定填充顏色，alpha指定透明度
ggplot(dss, aes(x = year, y = ss)) + 
geom_area(color = "green", fill = "green", alpha = 0.3)

正負分開堆積圖

dss[100:200,]$ss = -dss[100:200,]$ss
dss$tag <- dss$ss >= 0
# rle用來捕獲連續運行長度，比如
# dat <- c(1, 2, 2, 2, 3, 1, 4, 4, 1, 1)
# r <- rle(dat)
# r
#   lengths: int [1:6] 1 3 1 1 2 2
#   values : num [1:6] 1 2 3 1 4 1
cat.rle = rle(dss$tag)
# rep.int用來生成重複序列
# rep.int(1:5,2)
# 1 2 3 4 5 1 2 3 4 5
dss$group = rep.int(1:length(cat.rle$lengths), times=cat.rle$lengths)
# group的作用是把每一正負分段分爲不同的組
ggplot(dss, aes(x = year, y = ss, fill = tag, group = group)) + 
geom_area(alpha = 0.3) + 
scale_fill_manual(values = c('green','red'), guide = FALSE)

# 如果僅使用fill自動分組會出現奇怪的圖形，中間部分上方的紅色是由於ggplot將tag中被FALSE分隔的兩個TRUE看做同一group連接起來導致的
ggplot(dss, aes(x = year, y = ss, fill = tag)) + 
geom_area(alpha = 0.3) + 
scale_fill_manual(values = c('green','red'), guide = FALSE)

# 可以看到每一段連續的tag共享同一個group
dss[107:115,]

A data.frame: 9 × 4

	year	ss	tag	group
	<dbl>	<dbl>	<lgl>	<int>
107	1806	-28.1	FALSE	2
108	1807	-10.1	FALSE	2
109	1808	-8.1	FALSE	2
110	1809	-2.5	FALSE	2
111	1810	0.0	TRUE	3
112	1811	-1.4	FALSE	4
113	1812	-5.0	FALSE	4
114	1813	-12.2	FALSE	4
115	1814	-13.9	FALSE	4

堆積圖

堆積圖可以理解爲操作對象擴展至多個變量的面積圖。

平平無奇堆積圖

head(uspopage)

A data.frame: 6 × 3

Year	AgeGroup	Thousands
<int>	<fct>	<int>
1900	<5	9181
1900	5-14	16966
1900	15-24	14951
1900	25-34	12161
1900	35-44	9273
1900	45-54	6437

ggplot(uspopage, aes(x = Year, y = Thousands, fill = AgeGroup)) + 
geom_area(color = 'black', size = 0.2, alpha = 0.3) + 
scale_fill_brewer(palette = "YlOrRd")

順序逆轉堆積圖

library(plyr)
ggplot(uspopage, aes(x = Year, y = Thousands, fill = factor(uspopage$AgeGroup, levels = rev(levels(uspopage$AgeGroup))))) + 
geom_area(color = 'black', size = 0.2, alpha = 0.3) + 
scale_fill_brewer(palette = "YlOrRd", breaks = rev(levels(uspopage$AgeGroup))) + 
labs(fill="AgeGroup")

去除豎線堆積圖

ggplot(uspopage, aes(x = Year, y = Thousands, fill = AgeGroup)) + 
geom_area(alpha = 0.3) + 
scale_fill_brewer(palette = "YlOrRd") + 
geom_line(position = 'stack', color = 'black', size = 0.2)

# 折騰半天我感覺還是沒線的最好看
ggplot(uspopage, aes(x = Year, y = Thousands, fill = AgeGroup)) + 
geom_area(alpha = 0.3) + 
scale_fill_brewer(palette = "YlOrRd")

百分比比堆積圖

# 計算百分比，首先按年份分成多個獨立的dataframe，在每個frame內部計算present
uspopage_prop <- ddply(uspopage, "Year", transform, Persent = Thousands/sum(Thousands)*100)

ggplot(uspopage_prop, aes(x = Year, y = Persent, fill = AgeGroup)) + 
geom_area(alpha = 0.3) + 
scale_fill_brewer(palette = "YlOrRd")