r - 将 POSIXct 日期值更改为每周的第一天

Question

我想使用这些数据(如下)计算每周的平均 Dist，同时保留使用 POSIXct 时间类的优势。

    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

我的想法是使用 plyr 包按周平均 Dist 并希望首先创建一个包含日期的新 WeekDate 字段，不包括每周第一天的时间。正如 DoW(星期几)字段中所示，数据并不总是从一周的第一天开始。

虽然我似乎无法将这些点联系起来，但我想要每个连续周(DoW 1-7)的最短日期(不包括 h:m:s)。

第 1:10 行将是 2012-02-16， 第 11:38 行将是 2012-02-19， 第 39:50 行将是 2012-02-26

我怀疑lubridate包会有帮助，但无法获得正确的代码。

任何关于具体创建新日期列或更广泛地平均每周 Dist 的建议或替代方法将不胜感激。

Answer 1

使用plyr的好友，

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

还有ceiling_date、round_date选项。