我有一些看起来像这样的数据:
> ExampleData
# A tibble: 14,833 x 4
CountryCode Flow FuelType Value
<fct> <fct> <fct> <dbl>
1 ALB road coa 0
2 ALB services coa 3.3
3 ALB manufact coa 113.1
4 ALB mining coa 0
5 ALB road nga 0
6 ALB services nga 2.4
[...]
从代表国家,流量和燃料类型的每种组合的角度来看,数据是完整的。
我想将其转换为三维数组,其中国家表示一个维度,流动表示另一个维度,燃料类型表示第三个维度。因此,我可以引用数据X [a,b,c],其中b和c是我的因素CountryCode,Flow和FuelType的相应整数值。
所以我要寻找的是“宽”数据的多维形式。
答案 0 :(得分:1)
一个选项是xtabs
中的base R
out <- xtabs(Value ~ CountryCode + Flow + FuelType, data = ExampleData)
out
#, , FuelType = coa
# Flow
#CountryCode manufact mining road services
# ALB 113.1 0.0 0.0 3.3
#, , FuelType = nga
# Flow
#CountryCode manufact mining road services
# ALB 0.0 0.0 0.0 2.4
我们可以使用位置索引或键提取单个元素
out["ALB", "manufact", "coa"]
#[1] 113.1
或与tapply
tapply(ExampleData[['Value']], ExampleData[-4], FUN = I)
ExampleData <- structure(list(CountryCode = c("ALB", "ALB", "ALB", "ALB", "ALB",
"ALB"), Flow = c("road", "services", "manufact", "mining", "road",
"services"), FuelType = c("coa", "coa", "coa", "coa", "nga",
"nga"), Value = c(0, 3.3, 113.1, 0, 0, 2.4)),
class = "data.frame", row.names = c("1",
"2", "3", "4", "5", "6"))