[英]changing POSIXct date vaules to first day of each week
I want to calculate the average Dist
for each week using these data (below) while preserving the benefits of a using the POSIXct
time class. 我想使用这些数据(下面)计算每周的平均
Dist
,同时保留使用POSIXct
时间类的好处。
df <- structure(list(IndID = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), class = "factor", .Label = "AAA"),
Date = structure(c(1329436800, 1329458400, 1329480000, 1329501600,
1329523200, 1329544800, 1329566400, 1329588000, 1329609600,
1329631200, 1329652800, 1329674400, 1329696000, 1329717600,
1329739200, 1329760800, 1329782400, 1329804000, 1329825600,
1329847200, 1329868800, 1329890400, 1329912000, 1329933600,
1329955200, 1329976800, 1329998400, 1330020000, 1330041600,
1330063200, 1330084800, 1330106400, 1330128000, 1330149600,
1330171200, 1330192800, 1330214400, 1330236000, 1330257600,
1330279200, 1330300800, 1330322400, 1330344000, 1330365600,
1330387200, 1330408800, 1330430400, 1330452000, 1330473600,
1330495200), class = c("POSIXct", "POSIXt"), tzone = ""),
Dist = c(3.85567120344727, 52.2649622620809, 1043.61207930222,
1352.58506343616, 176.911523081261, 77.8266318470078, 50.3943567710686,
296.753649985307, 70.5826583995618, 166.394264991861, 251.745346701973,
295.70655057823, 44.6664731663839, 11.1539274078084, 124.578071475754,
757.728373470112, 83.0921234152083, 36.6820839851181, 29.1406161870034,
150.442928003814, 66.0957159105813, 2.23839297570488, 184.88312900824,
513.072526047611, 132.868335201626, 8.09274857805967, 284.479977841835,
479.358187122796, 297.273840894826, 4.00676616275076, 601.492189218489,
249.001525522847, 108.007775719885, 2.38435966274261, 604.365702677913,
1499.59076416313, 111.74722960012, 25.3528529967124, 280.057754683142,
428.157539641219, 70.0365608334965, 71.0886617898624, 265.823654634254,
380.247565078552, 188.857338305481, 9.24402933768915, 120.346786301264,
221.904294953242, 201.086079767386, 81.7857577639103), DoW = c(5,
5, 6, 6, 6, 6, 7, 7, 7, 7, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3,
3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 6, 6, 7, 7, 7, 7, 1,
1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3)), .Names = c("IndID", "Date",
"Dist", "DoW"), row.names = c(NA, -50L), class = "data.frame")
> head(df)
IndID Date Dist DoW
1 AAA 2012-02-16 17:00:00 3.855671 5
2 AAA 2012-02-16 23:00:00 52.264962 5
3 AAA 2012-02-17 05:00:00 1043.612079 6
4 AAA 2012-02-17 11:00:00 1352.585063 6
5 AAA 2012-02-17 17:00:00 176.911523 6
6 AAA 2012-02-17 23:00:00 77.826632 6
My thought was to use the plyr
package to average Dist
by week and wanted to first create a new WeekDate
field that contains the date, excluding the time, of the first day of each week. 我的想法是使用
plyr
包来按周平均Dist
,并且想要首先创建一个新的WeekDate
字段,其中包含每周第一天的日期,不包括时间。 As seen in the DoW (Day of Week) field, data does not always begin on the first day of the week. 如DoW(星期几)字段中所示,数据并不总是在一周的第一天开始。
While i cannot seem to connect the dots, I want the min Date excluding h:m:s) for each sequential week (DoW 1-7). 虽然我似乎无法连接点,但我希望每个连续周(DoW 1-7)的最小日期不包括h:m:s)。
Rows 1:10 would be 2012-02-16, Rows 11:38 would be 2012-02-19, Rows 39:50 would be 2012-02-26 行1:10将是2012-02-16,行11:38将是2012-02-19,行39:50将是2012-02-26
I suspect the lubridate
package will be helpful but can not get the code correct. 我怀疑
lubridate
包会有用,但无法正确获取代码。
Any suggestions or alternative methods on the specific creation of a new date column or more broadly averaging Dist for every week would be appreciated. 有关特定创建新日期列的任何建议或替代方法,或者更广泛地平均每周的Dist,我们将不胜感激。
Using bosom buddy of plyr
, 使用
plyr
bosom buddy,
library(lubridate)
library(dplyr)
df %>%
group_by(Week = floor_date(Date, unit="week")) %>%
summarize(WeeklyAveDist=mean(Dist))
#Source: local data frame [3 x 2]
#
# Week WeeklyAveDist
#1 2012-02-12 381.7755
#2 2012-02-19 252.1116
#3 2012-02-26 175.4097
There are also ceiling_date
, round_date
options. 还有
ceiling_date
, round_date
选项。
You could use strftime
with %W
format: 您可以使用
%W
格式的strftime
:
> strftime(as.Date("2015-01-08"), "%W")
[1] "01"
You could use this to define a new variable and then aggregate by this variable. 您可以使用它来定义新变量,然后通过此变量进行聚合。 Maybe thus
也许就这样
> df <- transform(df, week=strftime(Date, "%W"))
> aggregate(df$Dist, by=list(df$week), FUN=mean)
Group.1 x
1 07 319.8861
2 08 254.2861
3 09 161.0421
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.