如何减去数据集中某些行和每个面板中的值

Question

数据为：

   df <- structure(list(country = c("Poland", "Poland", "Poland", "Poland", "Poland", "Poland","Portugal", "Portugal", "Portugal", "Portugal", "Portugal", "Portugal", "Spain", "Spain", "Spain", "Spain", "Spain", "Spain"), Code = c("POL", "POL", "POL", "POL", "POL", "POL", "PRT", "PRT", "PRT", "PRT", "PRT", "PRT", "ESP", "ESP", "ESP", "ESP", "ESP", "ESP"), year = c(1950, 1951, 1952, 1953, 1954,1955, 1950, 1951, 1952, 1953, 1954, 1955, 1950, 1951, 1952, 1953, 1954, 1955), IV = c(1, 1, 1, 2, 3, 3, 1, 1, 1, 2, 2, 1, 1, 1, 2, 3, 4, 5)), class = c("tbl_df", "tbl", "data.frame"), row.names = c(NA, -18L))

我想从 1952 年减去 1955 年的 IV 值； 1954-1951； 1953-1950。对于每个国家来说也是如此。在结果数据集中（我们将其命名为“newdata”），新变量“newIV”必须显示这种差异； 1950、1951 和 1952 年的值应删除，如下所示：

Answer 1

您可以

merge

和

aggregate

d

subset

获得临时

iv55

列，您可以将其减去，最后

subset

加上所需的年份。

> merge(df, 
+       aggregate(cbind(iv55=IV) ~ Code, subset(df, year == 1955), I), all=TRUE,
+       sort=FALSE
+ ) |> transform(IV=IV - iv55, iv55=NULL) |> subset(!year %in% 1950:1952)
   Code  country year IV
4   POL   Poland 1953 -1
5   POL   Poland 1954  0
6   POL   Poland 1955  0
10  PRT Portugal 1953  1
11  PRT Portugal 1954  1
12  PRT Portugal 1955  0
16  ESP    Spain 1953 -2
17  ESP    Spain 1954 -1
18  ESP    Spain 1955  0