问题描述
我有一个 4 年的数据集，我想计算每天的最高温度（提供每小时 1 次测量）。如何在数据集中的附加列中添加此温度？我不知道如何在不删除其他列的情况下执行此操作。
我的数据集是这样的：
structure(list(DateTime = structure(c(1420070400,1420074000,1420077600,1420081200,1420084800,1420088400,1420092000,1420095600,1420099200,1420102800,1420106400,1420110000,1420113600,1420117200,1420120800,1420124400,1420128000,1420131600,1420135200,1420138800,1420142400,1420146000,1420149600,1420153200,1420156800,1420160400,1420164000,1420167600,1420171200,1420174800,1420178400,1420182000,1420185600,1420189200,1420192800,1420196400,1420200000,1420203600,1420207200,1420210800,1420214400,1420218000,1420221600,1420225200,1420228800,1420232400,1420236000,1420239600),tzone = "UTC",class = c("POSIXct","POSIXt")),Tmin = c(3.33733696,3.2377765,2.83953466,3.03865558,2.7399742,2.44129282,2.34173236,2.2421719,2.54085328,3.33733696,3.43689742,3.53645788,2.93909512,3.7355788,4.03426018,4.7311834,4.53206248,4.93030432,6.02546938,7.02107398,4.33294156,3.83513926,4.83074386,2.7399742),Tmax = c(3.77972493,3.38212841,3.18333015,2.98453189,2.48753624,2.38813711,2.28873798,2.78573363,3.48152754,4.27672058,4.17732145,2.88513276,2.6863345,3.28272928,3.6803258,3.97852319,3.87912406,4.6743171,4.87311536,4.47551884,4.77371623,5.76770753,6.86109796,7.85508926,8.74968143,5.37011101,6.16530405,4.97251449,5.17131275,2.98453189),Tmean = c(3.62254694166667,3.30742526,3.00888893,3.07523033666667,2.87620611666667,2.49474302833333,2.39523091833333,2.262548105,2.37864556666667,2.7103526,2.74352330333333,2.959132875,3.37376666666667,3.77181510666667,3.854741865,3.32401061166667,2.80986471,2.84303541333333,2.82645006166667,2.760108655,2.92596217166667,2.97571822666667,3.09181568833333,3.58937623833333,3.82157116166667,3.98742467833333,3.63913229333333,4.41864382166667,4.83327761333333,3.62254694166667,4.08693678833333,4.68400944833333,5.33083816333333,6.49181278,7.56986063833333,7.28790966,5.06547253666667,4.451814525,5.71230125166667,4.38547311833333,4.849862965,4.50157058,2.99230357833333,2.79327935833333
)),row.names = c(NA,-48L),class = c("tbl_df","tbl","data.frame"
))
解决方法
因为我在等待模型运行：
library(dplyr)
d <-
data.frame(date = sample(seq(as.Date('2021-01-01'),as.Date('2021-01-31'),by = 'day'),75,replace = T),value = rnorm(75))
d %>%
left_join(d %>%
group_by(date) %>%
summarize(max = max(value))) %>%
arrange(date)
dplyr() 库非常有用。它是 tidyverse 软件包的一部分，具有许多有用的数据管理功能。
我创建了一组数据来展示如何使用 summarize() 来计算每个 max() 日期的 group_by 值，然后 left_join() 将汇总数据集返回到原创。
编辑：这是@dario 的建议：
d %>%
group_by(date) %>%
mutate(max_val = max(value)) %>%
arrange(date)
更容易处理，不知道我是如何在第一轮比赛中错过的！
,编辑：
添加了日期-时间到日期-日期的转换：
在 mutate 中使用汇总：
dd 是问题中提供的数据结构：
library(dplyr)
dd_new <- dd %>%
mutate(dmy = as.POSIXct(as.Date(DateTime),"%d-%m-%Y")) %>%
group_by(dmy) %>%
mutate(max_temp = max(Tmax)) %>%
ungroup() %>%
as.data.frame()
str(dd_new)
现在返回：
'data.frame': 48 obs. of 6 variables:
$ DateTime: POSIXct,format: "2015-01-01 00:00:00" "2015-01-01 01:00:00" "2015-01-01 02:00:00" ...
$ Tmin : num 3.34 3.24 2.84 3.04 2.74 ...
$ Tmax : num 3.78 3.38 3.18 3.18 2.98 ...
$ Tmean : num 3.62 3.31 3.01 3.08 2.88 ...
$ dmy : POSIXct,format: "2015-01-01 01:00:00" "2015-01-01 01:00:00" "2015-01-01 01:00:00" ...
$ max_temp: num 4.28 4.28 4.28 4.28 4.28 ...