我会尽我所能，使我的第一个问题尽可能具体和可重复。 我有一个庞大的数据集，具有不同年份（2010-2019年）不同站点（例如Carb_Ben，Carb_uR，OW_Kessin）的不同参数（例如TP和O2）的年度平均值或最小值。每个电台都属于特定类型（例如：11_tl_ba，14_ka，23）。
data<-structure(list(Station = c("Carb_Ben", "Carb_Ben", "Carb_uR",
"Carb_uR", "Laak_GK", "Laak_GK", "Laak_GK", "OW_Kessin", "OW_Kessin",
"OW_Kessin"), Typ = c("11_tl_ba", "11_tl_ba", "11_tl_ba", "11_tl_ba",
"23", "23", "23", "23", "23", "23"), Jahr = structure(c(4L, 7L,
4L, 7L, 2L, 5L, 8L, 3L, 4L, 5L), .Label = c("2010", "2013", "2014",
"2015", "2016", "2017", "2018", "2019"), class = "factor"), O2_min = c(1.44,
1.53, 8, 6.7, 2.7, 1.79, 1.4, 4.3, 4.7675, 4.015), TP_mean = c(0.513333333333333,
0.1625, 0.148333333333333, 0.124166666666667, 0.155, 0.28, 0.175833333333333,
0.0954166666666667, 0.0929166666666667, 0.0970833333333333)), row.names = c(NA,
-10L), groups = structure(list(Station = c("Carb_Ben", "Carb_uR",
"Laak_GK", "OW_Kessin"), Typ = c("11_tl_ba", "11_tl_ba", "23",
"23"), .rows = list(1:2, 3:4, 5:7, 8:10)), row.names = c(NA,
-4L), class = c("tbl_df", "tbl", "data.frame"), .drop = TRUE), class = c("grouped_df",
"tbl_df", "tbl", "data.frame"))
#so the data look like this:
# A tibble: 10 x 5
# Groups: Station, Typ [4]
Station Typ Jahr O2_min TP_mean
<chr> <chr> <fct> <dbl> <dbl>
1 Carb_Ben 11_tl_ba 2015 1.44 0.513
2 Carb_Ben 11_tl_ba 2018 1.53 0.162
3 Carb_uR 11_tl_ba 2015 8 0.148
4 Carb_uR 11_tl_ba 2018 6.7 0.124
5 Laak_GK 23 2013 2.7 0.155
6 Laak_GK 23 2016 1.79 0.28
7 Laak_GK 23 2019 1.4 0.176
8 OW_Kessin 23 2014 4.3 0.0954
9 OW_Kessin 23 2015 4.77 0.0929
10 OW_Kessin 23 2016 4.01 0.0971
然后，我为每个参数和类型指定了状态（_g）或状态（_sg）很好的条件。
criteria<-structure(list(Typ = structure(c(10L, 5L, 24L), .Label = c("10",
"11_alp", "11_mg_ba", "11_mg_br", "11_tl_ba", "11_tl_br", "12_mg_ba",
"12_tl_ba", "12_tl_br", "14_ka", "14_si", "15", "15 g", "16_ka",
"16_si", "17", "18", "19", "19_ba", "2.1", "2.2", "20", "22",
"23", "3.1", "3.2", "4", "5", "5.1", "6", "6 K", "7", "9", "9.1",
"9.1 K", "9.2", "B1", "B2a", "B2b", "B3a", "B3b", "Subtyp 21 N"
), class = "factor"), O2_g = c(7L, 6L, 4L), O2_sg = 9:7, TP_g = c(0.1,
0.15, 0.1), TP_sg = c(0.05, 0.05, 0.05)), row.names = c(24L,
28L, 37L), class = "data.frame")
#This looks like this:
Typ O2_g O2_sg TP_g TP_sg
24 11_tl_ba 8 8 0.10 0.05
28 14_ka 7 9 0.10 0.05
27 23 6 8 0.15 0.05
现在，我想比较每个台站和每年的每个参数（如果它取决于类型，是否符合标准）。查明每年每个站的每个参数的状态是“好”（肠），“很好”（肠）还是“不好”（肠）
我目前的解决方案
data_criteria<-data%>%
mutate(GW_O2_g=case_when(Typ=='11_tl_ba' ~criteria$O2_g[which(criteria$Typ=='11_tl_ba')],
Typ=='14_ka' ~criteria$O2_g[which(criteria$Typ=='14_ka')],
Typ=='23' ~criteria$O2_g[which(criteria$Typ=='23')]),
GW_O2_sg=case_when(Typ=='11_tl_ba' ~criteria$O2_sg[which(criteria$Typ=='11_tl_ba')],
Typ=='14_ka' ~criteria$O2_sg[which(criteria$Typ=='14_ka')],
Typ=='23' ~criteria$O2_sg[which(criteria$Typ=='23')]),
GW_TP_g=case_when(Typ=='11_tl_ba' ~criteria$TP_g[which(criteria$Typ=='11_tl_ba')],
Typ=='14_ka' ~criteria$TP_g[which(criteria$Typ=='14_ka')],
Typ=='23' ~criteria$TP_g[which(criteria$Typ=='23')]),
GW_TP_sg=case_when(Typ=='11_tl_ba' ~criteria$TP_sg[which(criteria$Typ=='11_tl_ba')],
Typ=='14_ka' ~criteria$TP_sg[which(criteria$Typ=='14_ka')],
Typ=='23' ~criteria$TP_sg[which(criteria$Typ=='23')]))%>%
mutate(GK_O2=case_when(O2_min<GW_O2_g ~'nicht gut',
O2_min>=GW_O2_g & O2_min<GW_O2_sg~'gut',
O2_min>GW_O2_sg ~ 'sehr gut'),
GK_TP=case_when(TP_mean>GW_TP_g ~ 'nicht gut',
TP_mean<=GW_TP_g & TP_mean>GW_TP_sg ~ 'gut',
TP_mean<=GW_TP_sg ~ 'sehr gut'))
#to get this:
# A tibble: 10 x 11
# Groups: Station, Typ [4]
Station Typ Jahr O2_min TP_mean GW_O2_g GW_O2_sg GW_TP_g GW_TP_sg GK_O2 GK_TP
<chr> <chr> <fct> <dbl> <dbl> <int> <int> <dbl> <dbl> <chr> <chr>
1 Carb_Ben 11_tl_ba 2015 1.44 0.513 8 8 0.1 0.05 nicht gut nicht gut
2 Carb_Ben 11_tl_ba 2018 1.53 0.162 8 8 0.1 0.05 nicht gut nicht gut
3 Carb_uR 11_tl_ba 2015 8 0.148 8 8 0.1 0.05 sehr gut nicht gut
4 Carb_uR 11_tl_ba 2018 6.7 0.124 8 8 0.1 0.05 nicht gut nicht gut
5 Laak_GK 23 2013 2.7 0.155 6 8 0.15 0.05 nicht gut nicht gut
6 Laak_GK 23 2016 1.79 0.28 6 8 0.15 0.05 nicht gut nicht gut
7 Laak_GK 23 2019 1.4 0.176 6 8 0.15 0.05 nicht gut nicht gut
8 OW_Kessin 23 2014 4.3 0.0954 6 8 0.15 0.05 nicht gut gut
9 OW_Kessin 23 2015 4.77 0.0929 6 8 0.15 0.05 nicht gut gut
10 OW_Kessin 23 2016 4.01 0.0971 6 8 0.15 0.05 nicht gut gut
由于我拥有更多的工作站和类型以及更多的参数，因此我的解决方案意味着给我更多的键入内容以及更多的包含错误的机会。 我很确定还有一种更优雅的方法可以实现相同的结果，但是到目前为止，我对R的了解还不深，以至于该怎么做？
因此，如果有人有一个很好的简单解决方案，我将非常高兴；）。
答案 0 :(得分：0)
您可以使用criteria从match中反复选择。 within有助于保持代码简洁。
dat <- within(dat, {
GW_O2_g <- criteria[match(dat$Typ, criteria$Typ), "O2_g"]
GW_O2_sg <- criteria[match(dat$Typ, criteria$Typ), "O2_sg"]
GW_TP_g <- criteria[match(dat$Typ, criteria$Typ), "TP_g"]
GW_TP_sg <- criteria[match(dat$Typ, criteria$Typ), "TP_sg"]
GK_O2 <- NA
GK_O2[O2_min < GW_O2_g] <- "nicht gut"
GK_O2[O2_min >= GW_O2_g & O2_min < GW_O2_sg] <- "gut"
GK_O2[O2_min > GW_O2_sg] <- "sehr gut"
GK_TP <- NA
GK_TP[TP_mean > GW_TP_g] <- "nicht gut"
GK_TP[TP_mean <= GW_TP_g & TP_mean > GW_TP_sg] <- "gut"
GK_TP[TP_mean <= GW_TP_sg] <- "sehr gut"
})
# # A tibble: 10 x 11
# # Groups: Station, Typ [4]
# Station Typ Jahr O2_min TP_mean GK_TP GK_O2 GW_TP_sg GW_TP_g GW_O2_sg GW_O2_g
# <chr> <chr> <fct> <dbl> <dbl> <chr> <chr> <dbl> <dbl> <int> <int>
# 1 Carb_Ben 11_tl_ba 2015 1.44 0.513 nicht gut nicht gut 0.05 0.15 8 6
# 2 Carb_Ben 11_tl_ba 2018 1.53 0.162 nicht gut nicht gut 0.05 0.15 8 6
# 3 Carb_uR 11_tl_ba 2015 8 0.148 gut NA 0.05 0.15 8 6
# 4 Carb_uR 11_tl_ba 2018 6.7 0.124 gut gut 0.05 0.15 8 6
# 5 Laak_GK 23 2013 2.7 0.155 nicht gut nicht gut 0.05 0.1 7 4
# 6 Laak_GK 23 2016 1.79 0.28 nicht gut nicht gut 0.05 0.1 7 4
# 7 Laak_GK 23 2019 1.4 0.176 nicht gut nicht gut 0.05 0.1 7 4
# 8 OW_Kessin 23 2014 4.3 0.0954 gut gut 0.05 0.1 7 4
# 9 OW_Kessin 23 2015 4.77 0.0929 gut gut 0.05 0.1 7 4
# 10 OW_Kessin 23 2016 4.01 0.0971 gut gut 0.05 0.1 7 4