Question

我有两个数据框。 df1是一个数据帧，在每组单元中包括多个位置。 df2包括所有位置的每日最高温度（df $ tmax）的每日观测值。对于df1中的每个单位，我想计算平均每日最高费用。每个单元内所有位置的温度。

以下代码生成每个数据帧的示例。我需要将其扩展到约240个单位和8年的每日数据。

R中的这种查找/匹配练习似乎总能吸引我。一定有一种明显的方法可以做到这一点，但目前我受阻，没有加入真正的蛮力等。

df1 <- 
structure(list(unitID = c("98008", "98008", "98008", "98008", 
"98065", "98065", "98065", "98065", "98146", "98146", "98146", 
"98146", "98584", "98584", "98584"), locationID = c("USW00094290", "USW00094248", 
"USW00024234", "USC00454169", "USC00458508", "USS0021B60S", "USR0000WFTA", 
"USC00451233", "USW00024234", "USW00024233", "USW00094248", "USC00454169", 
"USW00094227", "USC00451939", "USC00455086")), class = "data.frame", row.names = c(NA, 
-15L))

df1
unitID  locationID
1   98008 USW00094290
2   98008 USW00094248
3   98008 USW00024234
4   98008 USC00454169
5   98065 USC00458508
6   98065 USS0021B60S
7   98065 USR0000WFTA
8   98065 USC00451233
9   98146 USW00024234
10  98146 USW00024233
11  98146 USW00094248
12  98146 USC00454169
13  98584 USW00094227
14  98584 USC00451939
15  98584 USC00455086

df2 <- 
structure(list(id = c("USW00094290", "USW00094290", "USW00094248", 
"USW00094248", "USW00024234", "USW00024234", "USC00454169", "USC00454169", 
"USC00458508", "USC00458508", "USS0021B60S", "USS0021B60S", "USR0000WFTA", 
"USR0000WFTA", "USC00451233", "USC00451233", "USW00024233", "USW00024233", 
"USW00094227", "USW00094227", "USC00451939", "USC00451939", "USC00455086", 
"USC00455086"), date = structure(c(17167, 17168, 17167, 17168, 
17167, 17168, 17167, 17168, 17167, 17168, 17167, 17168, 17167, 
17168, 17167, 17168, 17167, 17168, 17167, 17168, 17167, 17168, 
17167, 17168), class = "Date"), tmax = c(28, 28, 28, 28, 33, 
28, 33, 28, -11, -28, -17, -50, 11, -17, 0, -11, 28, 11, 44, 
33, 50, 39, 39, 28)), row.names = c(NA, -24L), class = c("tbl_df", 
"tbl", "data.frame"))

df2
# A tibble: 24 x 3
   id          date        tmax
   <chr>       <date>     <dbl>
 1 USW00094290 2017-01-01    28
 2 USW00094290 2017-01-02    28
 3 USW00094248 2017-01-01    28
 4 USW00094248 2017-01-02    28
 5 USW00024234 2017-01-01    33
 6 USW00024234 2017-01-02    28
 7 USC00454169 2017-01-01    33
 8 USC00454169 2017-01-02    28
 9 USC00458508 2017-01-01   -11
10 USC00458508 2017-01-02   -28
# ... with 14 more rows

输出应包括unitID，日期和最大平均值。温度

unitID  date         avg_temp
98008   2009-01-01   30.5
98008   2009-01-02   ...
98008   2009-01-03   ...

Answer 1

我们可以使用n=5000; t=15000; lambda=0.8; sigmae1=0.05; sigmae2=0.1; sigmaz= 0.013; n_lambda= trunc(Int, lambda*n) eshocks1=rand(Normal(0.0,sigmae1), n_lambda, t); eshocks2=rand(Normal(0.0,sigmae2), n - n_lambda, t); zshocks =rand(Normal(0.0, sigmaz),1, t);，left_join group_by和unitID并取locationID中的mean。

tmax

在基数R中，我们可以使用library(dplyr) df1 %>% left_join(df2, by = c("locationID" = "id")) %>% group_by(unitID, locationID) %>% summarise(tmx = mean(tmax, na.rm = TRUE)) # unitID locationID tmx # <chr> <chr> <dbl> # 1 98008 USC00454169 30.5 # 2 98008 USW00024234 30.5 # 3 98008 USW00094248 28 # 4 98008 USW00094290 28 # 5 98065 USC00451233 -5.5 # 6 98065 USC00458508 -19.5 # 7 98065 USR0000WFTA -3 # 8 98065 USS0021B60S -33.5 # 9 98146 USC00454169 30.5 #10 98146 USW00024233 19.5 #11 98146 USW00024234 30.5 #12 98146 USW00094248 28 #13 98584 USC00451939 44.5 #14 98584 USC00455086 33.5 #15 98584 USW00094227 38.5和merge

aggregate

Answer 2

我们可以使用data.table联接

library(data.table)
setDT(df1)[setDT(df2), on = .(locationID = id)][,
    .(tmx = mean(tmax, na.rm = TRUE)), .(unitID, locationID)]
#.   unitID  locationID   tmx
# 1:  98008 USW00094290  28.0
# 2:  98008 USW00094248  28.0
# 3:  98146 USW00094248  28.0
# 4:  98008 USW00024234  30.5
# 5:  98146 USW00024234  30.5
# 6:  98008 USC00454169  30.5
# 7:  98146 USC00454169  30.5
# 8:  98065 USC00458508 -19.5
# 9:  98065 USS0021B60S -33.5
#10:  98065 USR0000WFTA  -3.0
#11:  98065 USC00451233  -5.5
#12:  98146 USW00024233  19.5
#13:  98584 USW00094227  38.5
#14:  98584 USC00451939  44.5
#15:  98584 USC00455086  33.5

跨R中的数据框汇总数据

2 个答案: