计算数年数月的中值
Calculating median value for months for several years
我最近开始使用 R 并尝试解决以下问题。我有 data.frame 包括一年中特定月份的到达和离开。我必须找出这些年来每个月的中位数是多少。我的结果应该保存在 .csv 中。以下只是示例，整个观察包括截至 2017 年的日期（总共 1548 个观察）：
#dput output assigned to the flights variable
flights <- structure(list(X = 1:163, ReportPeriod = structure(c(1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L, 3L, 3L, 3L, 3L, 3L,
3L, 3L, 3L, 3L, 3L, 3L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L,
5L, 5L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 8L, 8L, 8L,
8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 9L, 9L, 9L, 9L, 9L, 9L, 9L,
9L, 9L, 9L, 9L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L,
10L, 10L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 11L,
12L, 12L, 12L, 12L, 12L, 12L, 12L, 12L, 12L, 12L, 12L, 13L, 13L,
13L, 13L, 13L, 13L, 13L, 13L, 13L, 13L, 13L, 14L, 14L, 14L, 14L,
14L, 14L, 14L, 14L, 14L, 14L, 14L, 14L, 15L, 15L, 15L, 15L, 15L,
15L, 15L, 15L, 15L, 15L, 15L, 15L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 6L), .Label = c("01/01/2006 12:00:00 AM", "01/01/2007 12:00:00 AM",
"02/01/2006 12:00:00 AM", "02/01/2007 12:00:00 AM", "03/01/2006 12:00:00 AM",
"03/01/2007 12:00:00 AM", "04/01/2006 12:00:00 AM", "05/01/2006 12:00:00 AM",
"06/01/2006 12:00:00 AM", "07/01/2006 12:00:00 AM", "08/01/2006 12:00:00 AM",
"09/01/2006 12:00:00 AM", "10/01/2006 12:00:00 AM", "11/01/2006 12:00:00 AM",
"12/01/2006 12:00:00 AM"), class = "factor"), FlightType = structure(c(1L,
1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L,
2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L,
3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L,
1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L,
2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L,
1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L,
2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L,
3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L,
1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L,
2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L,
3L, 1L), .Label = c("Charter", "Commuter", "Scheduled"), class = "factor"),
Arrival_Departure = structure(c(1L, 1L, 2L, 2L, 1L, 1L, 2L,
2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L,
2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L,
2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 2L, 2L, 1L, 1L,
2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L,
1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L,
2L, 1L, 1L, 2L, 2L, 1L), .Label = c("Arrival", "Departure"
), class = "factor"), Domestic_International = structure(c(1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L,
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L,
1L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L,
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L), .Label = c("Domestic",
"International"), class = "factor"), FlightOpsCount = c(57L,
299L, 62L, 5L, 5996L, 31L, 5995L, 31L, 13695L, 4026L, 13666L,
4085L, 22L, 101L, 21L, 100L, 5525L, 28L, 5525L, 28L, 12308L,
3381L, 12322L, 3398L, 51L, 4L, 47L, 5L, 6217L, 31L, 6216L,
30L, 13925L, 3952L, 13963L, 3961L, 57L, 2L, 52L, 5885L, 31L,
5886L, 31L, 13420L, 3884L, 13461L, 3897L, 26L, 5L, 24L, 3L,
6089L, 41L, 6089L, 41L, 14126L, 3968L, 14138L, 3984L, 18L,
1L, 18L, 5980L, 39L, 5980L, 39L, 14208L, 4030L, 14211L, 4028L,
16L, 2L, 14L, 6092L, 39L, 6092L, 39L, 14866L, 4483L, 14883L,
4506L, 15L, 1L, 16L, 6134L, 40L, 6134L, 40L, 15243L, 4303L,
15272L, 4328L, 24L, 1L, 18L, 5903L, 37L, 5903L, 37L, 13832L,
3823L, 13815L, 3865L, 48L, 285L, 50L, 6135L, 40L, 6135L,
40L, 14292L, 3605L, 14332L, 3898L, 48L, 3L, 50L, 2L, 5772L,
39L, 5772L, 39L, 13855L, 3706L, 13855L, 3718L, 46L, 6L, 44L,
3L, 6144L, 40L, 6144L, 40L, 14417L, 4086L, 14474L, 4103L,
38L, 3L, 42L, 3L, 6081L, 94L, 6081L, 94L, 14140L, 4301L,
14165L, 4308L, 14L, 7L, 16L, 5L, 5470L, 83L, 5470L, 83L,
12734L, 3779L, 12768L, 3783L, 33L)), .Names = c("X", "ReportPeriod",
"FlightType", "Arrival_Departure", "Domestic_International",
"FlightOpsCount"), class = "data.frame", row.names = c(NA, -163L
))
这是我应该得到的预期输出：
month,Arrival_Departure,FlightOpsCount
January,Arrival,23152
January,Departure,23173
February,Arrival,20849
February,Departure,20878
March,Arrival,23981.5
March,Departure,24005
April,Arrival,23393
April,Departure,23451.5
May,Arrival,24342.5
May,Departure,24376
June,Arrival,24622.5
June,Departure,24667
July,Arrival,25795.5
July,Departure,25837
August,Arrival,25573.5
August,Departure,25600
September,Arrival,23306
September,Departure,23315
October,Arrival,23965
October,Departure,23990
November,Arrival,22379
November,Departure,22361
December,Arrival,23594
December,Departure,23579
我决定分几个步骤来做，我尝试做的第一件事是从字符串接收正确的日期格式：
library(dbplyr)
step_1 = as_tibble(flights)
step_2 = step_1 %>%
transmute(
date_format = as.POSIXct(strptime(ReportPeriod, format = "%m/%d/%Y")),
even_new_date = as.Date(date_format, format = "%Y"),
Arrival_Departure,
FlightOpsCount)
这对我来说实际上很棘手..我不明白如何正确地做到这一点以及为什么有两种方法来获取日期格式，例如 2006-01-01 和 2005-12-31？在这种情况下，哪一个是正确的？
现在，假设 2006-01-01 是正确的，我可以在 months() 函数中使用 as.POSIXct 来获取月份：
step2 = step_1 %>%
transmute(
month = months(as.POSIXct(strptime(ReportPeriod, format = "%m/%d/%Y"))),
Arrival_Departure,
FlightOpsCount)
下一步需要分组操作：
step_3 = step_2 %>%
group_by(month, Arrival_Departure) %>%
summarize(median = median(FlightOpsCount))
将其写入 csv 时，我得到的值小得离谱..
"","month","Arrival_Departure","median"
"1","April","Arrival",102.5
"2","April","Departure",3061
"3","August","Arrival",1412.5
"4","August","Departure",3667.5
"5","December","Arrival",102
"6","December","Departure",1738
"7","February","Arrival",116
"8","February","Departure",116
"9","January","Arrival",284
"10","January","Departure",1708
"11","July","Arrival",95.5
"12","July","Departure",3571
"13","June","Arrival",119
"14","June","Departure",3292
"15","March","Arrival",115
"16","March","Departure",1759
"17","May","Arrival",1609.5
"18","May","Departure",3121
"19","November","Arrival",93.5
"20","November","Departure",93.5
"21","October","Arrival",2359
"22","October","Departure",2756
"23","September","Arrival",1228
"24","September","Departure",3187.5
有人可以指导我解决问题并告诉我解决问题的正确方法吗？
如有任何帮助，我将不胜感激。
你快到了，不过我建议使用 dplyr:
# Step 1: Convert dates using as.Date function
flights$ReportPeriod <- as.Date(flights$ReportPeriod, "%m/%d/%Y")
# Step 2: Use dplyr to summarize information
require(dplyr)
flights <- flights %>%
group_by(ReportPeriod, Arrival_Departure) %>%
summarise(FlightOpsCount = median(FlightOpsCount)) %>%
as.data.frame()
# Step 3: Convert date to string for month name
flights <- flights %>%
mutate(ReportPeriod = months(ReportPeriod)) %>%
rename(month = ReportPeriod) # If you need to rename the column to be "months"
# Alternate Step 3: If you want to add in year as well
require(lubridate)
flights <- flights %>%
mutate(ReportPeriod = paste(months(ReportPeriod),
year(ReportPeriod),
sep = " ")) %>%
rename(month = ReportPeriod) # If you need to rename the column to be "months"
# Step 4: Write to csv
write.csv(flights, "file_name.csv", row.names = FALSE)
干杯。
这是一个 data.table 方法：
library(data.table)
library(lubridate)
dat <- fread("sample_data.txt", col.names = c("dte", "flight", "typ1","typ2","flt_count"))
dat$dte <- as.POSIXct(strptime(dat$dte, format = "%m/%d/%Y %H:%M:%S"), tz = "GMT")
new_dat <- dat[, sum(flt_count), by = list(month(dte),typ1)]
为了方便起见，我重命名了这些列。您也可以根据其他变量将 by 参数编辑为 group/perform 操作。上面代码片段的输出是：
> new_dat
month typ1 V1
1: 1 Arrival 24104
2: 1 Departure 23844
3: 2 Arrival 21365
4: 2 Departure 21394
5: 3 Arrival 24180
6: 3 Departure 24222
这似乎是您要查找的内容。 data.table 对于大型数据集非常有用。
写结果，使用write.csv(new_dat, "new_file.csv", row_names = F)
希望这对您有所帮助。
我相信这要简单得多。请注意，months 的格式与您的略有不同。
library(zoo)
months <- as.yearmon(flights$ReportPeriod, "%m/%d/%Y %H:%M:%S")
agg <- aggregate(FlightOpsCount ~ months + Arrival_Departure, flights, median)
无论如何，我无法得到甚至不接近你问题中预期值的中位数。由于这些是中位数，您的预期似乎有些错误。
如果您想要其他日期格式，可以使用函数 format.Date。
format(as.Date(months), "%Y %B") # or "%B %Y"
有关可能的格式的长列表，请参阅帮助页面 ?strptime。
谢谢大家的帮助！我确实收到了特定月份的正确值，这是我的代码：
#summarize Arrival & Departures through the years
step_1 <- flights %>%
group_by(ReportPeriod, Arrival_Departure) %>%
summarise(sum = sum(FlightOpsCount)) %>%
arrange(ReportPeriod) %>%
ungroup()
#modify date format in ReportPeriod column to receive months
step_2 <- step_1 %>%
transmute(month = months(as.Date(ReportPeriod,"%m/%d/%Y")),
Arrival_Departure,
sum) %>%
group_by(month, Arrival_Departure) %>%
summarise(FlightOpsCount = median(sum)) %>%
write.csv(., "flights_output.csv", row.names = FALSE, quote = FALSE)
但是，我得到的月份是按字母顺序而不是按时间顺序排列的。我在这里的某个地方找到了解决方案，但它不能正常工作，我只得到 NAs。显然我在将任何内容写入 .csv 之前调用它并在 step_2.
的末尾添加 ungroup()
step_3 <- step_2 %>%
mutate(month = factor(month.name[month], levels = month.name)) %>%
arrange(month)
我最近开始使用 R 并尝试解决以下问题。我有 data.frame 包括一年中特定月份的到达和离开。我必须找出这些年来每个月的中位数是多少。我的结果应该保存在 .csv 中。以下只是示例，整个观察包括截至 2017 年的日期（总共 1548 个观察）：
#dput output assigned to the flights variable
flights <- structure(list(X = 1:163, ReportPeriod = structure(c(1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L, 3L, 3L, 3L, 3L, 3L,
3L, 3L, 3L, 3L, 3L, 3L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L,
5L, 5L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 8L, 8L, 8L,
8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 9L, 9L, 9L, 9L, 9L, 9L, 9L,
9L, 9L, 9L, 9L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L,
10L, 10L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 11L,
12L, 12L, 12L, 12L, 12L, 12L, 12L, 12L, 12L, 12L, 12L, 13L, 13L,
13L, 13L, 13L, 13L, 13L, 13L, 13L, 13L, 13L, 14L, 14L, 14L, 14L,
14L, 14L, 14L, 14L, 14L, 14L, 14L, 14L, 15L, 15L, 15L, 15L, 15L,
15L, 15L, 15L, 15L, 15L, 15L, 15L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 6L), .Label = c("01/01/2006 12:00:00 AM", "01/01/2007 12:00:00 AM",
"02/01/2006 12:00:00 AM", "02/01/2007 12:00:00 AM", "03/01/2006 12:00:00 AM",
"03/01/2007 12:00:00 AM", "04/01/2006 12:00:00 AM", "05/01/2006 12:00:00 AM",
"06/01/2006 12:00:00 AM", "07/01/2006 12:00:00 AM", "08/01/2006 12:00:00 AM",
"09/01/2006 12:00:00 AM", "10/01/2006 12:00:00 AM", "11/01/2006 12:00:00 AM",
"12/01/2006 12:00:00 AM"), class = "factor"), FlightType = structure(c(1L,
1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L,
2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L,
3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L,
1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L,
2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L,
1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L,
2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L,
3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L,
1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L,
2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L,
3L, 1L), .Label = c("Charter", "Commuter", "Scheduled"), class = "factor"),
Arrival_Departure = structure(c(1L, 1L, 2L, 2L, 1L, 1L, 2L,
2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L,
2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L,
2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 2L, 2L, 1L, 1L,
2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L,
1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L,
1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L,
2L, 1L, 1L, 2L, 2L, 1L), .Label = c("Arrival", "Departure"
), class = "factor"), Domestic_International = structure(c(1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L,
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L,
1L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L,
1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L,
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L), .Label = c("Domestic",
"International"), class = "factor"), FlightOpsCount = c(57L,
299L, 62L, 5L, 5996L, 31L, 5995L, 31L, 13695L, 4026L, 13666L,
4085L, 22L, 101L, 21L, 100L, 5525L, 28L, 5525L, 28L, 12308L,
3381L, 12322L, 3398L, 51L, 4L, 47L, 5L, 6217L, 31L, 6216L,
30L, 13925L, 3952L, 13963L, 3961L, 57L, 2L, 52L, 5885L, 31L,
5886L, 31L, 13420L, 3884L, 13461L, 3897L, 26L, 5L, 24L, 3L,
6089L, 41L, 6089L, 41L, 14126L, 3968L, 14138L, 3984L, 18L,
1L, 18L, 5980L, 39L, 5980L, 39L, 14208L, 4030L, 14211L, 4028L,
16L, 2L, 14L, 6092L, 39L, 6092L, 39L, 14866L, 4483L, 14883L,
4506L, 15L, 1L, 16L, 6134L, 40L, 6134L, 40L, 15243L, 4303L,
15272L, 4328L, 24L, 1L, 18L, 5903L, 37L, 5903L, 37L, 13832L,
3823L, 13815L, 3865L, 48L, 285L, 50L, 6135L, 40L, 6135L,
40L, 14292L, 3605L, 14332L, 3898L, 48L, 3L, 50L, 2L, 5772L,
39L, 5772L, 39L, 13855L, 3706L, 13855L, 3718L, 46L, 6L, 44L,
3L, 6144L, 40L, 6144L, 40L, 14417L, 4086L, 14474L, 4103L,
38L, 3L, 42L, 3L, 6081L, 94L, 6081L, 94L, 14140L, 4301L,
14165L, 4308L, 14L, 7L, 16L, 5L, 5470L, 83L, 5470L, 83L,
12734L, 3779L, 12768L, 3783L, 33L)), .Names = c("X", "ReportPeriod",
"FlightType", "Arrival_Departure", "Domestic_International",
"FlightOpsCount"), class = "data.frame", row.names = c(NA, -163L
))
这是我应该得到的预期输出：
month,Arrival_Departure,FlightOpsCount
January,Arrival,23152
January,Departure,23173
February,Arrival,20849
February,Departure,20878
March,Arrival,23981.5
March,Departure,24005
April,Arrival,23393
April,Departure,23451.5
May,Arrival,24342.5
May,Departure,24376
June,Arrival,24622.5
June,Departure,24667
July,Arrival,25795.5
July,Departure,25837
August,Arrival,25573.5
August,Departure,25600
September,Arrival,23306
September,Departure,23315
October,Arrival,23965
October,Departure,23990
November,Arrival,22379
November,Departure,22361
December,Arrival,23594
December,Departure,23579
我决定分几个步骤来做，我尝试做的第一件事是从字符串接收正确的日期格式：
library(dbplyr)
step_1 = as_tibble(flights)
step_2 = step_1 %>%
transmute(
date_format = as.POSIXct(strptime(ReportPeriod, format = "%m/%d/%Y")),
even_new_date = as.Date(date_format, format = "%Y"),
Arrival_Departure,
FlightOpsCount)
这对我来说实际上很棘手..我不明白如何正确地做到这一点以及为什么有两种方法来获取日期格式，例如 2006-01-01 和 2005-12-31？在这种情况下，哪一个是正确的？
现在，假设 2006-01-01 是正确的，我可以在 months() 函数中使用 as.POSIXct 来获取月份：
step2 = step_1 %>%
transmute(
month = months(as.POSIXct(strptime(ReportPeriod, format = "%m/%d/%Y"))),
Arrival_Departure,
FlightOpsCount)
下一步需要分组操作：
step_3 = step_2 %>%
group_by(month, Arrival_Departure) %>%
summarize(median = median(FlightOpsCount))
将其写入 csv 时，我得到的值小得离谱..
"","month","Arrival_Departure","median"
"1","April","Arrival",102.5
"2","April","Departure",3061
"3","August","Arrival",1412.5
"4","August","Departure",3667.5
"5","December","Arrival",102
"6","December","Departure",1738
"7","February","Arrival",116
"8","February","Departure",116
"9","January","Arrival",284
"10","January","Departure",1708
"11","July","Arrival",95.5
"12","July","Departure",3571
"13","June","Arrival",119
"14","June","Departure",3292
"15","March","Arrival",115
"16","March","Departure",1759
"17","May","Arrival",1609.5
"18","May","Departure",3121
"19","November","Arrival",93.5
"20","November","Departure",93.5
"21","October","Arrival",2359
"22","October","Departure",2756
"23","September","Arrival",1228
"24","September","Departure",3187.5
有人可以指导我解决问题并告诉我解决问题的正确方法吗？
如有任何帮助，我将不胜感激。
你快到了，不过我建议使用 dplyr:
# Step 1: Convert dates using as.Date function
flights$ReportPeriod <- as.Date(flights$ReportPeriod, "%m/%d/%Y")
# Step 2: Use dplyr to summarize information
require(dplyr)
flights <- flights %>%
group_by(ReportPeriod, Arrival_Departure) %>%
summarise(FlightOpsCount = median(FlightOpsCount)) %>%
as.data.frame()
# Step 3: Convert date to string for month name
flights <- flights %>%
mutate(ReportPeriod = months(ReportPeriod)) %>%
rename(month = ReportPeriod) # If you need to rename the column to be "months"
# Alternate Step 3: If you want to add in year as well
require(lubridate)
flights <- flights %>%
mutate(ReportPeriod = paste(months(ReportPeriod),
year(ReportPeriod),
sep = " ")) %>%
rename(month = ReportPeriod) # If you need to rename the column to be "months"
# Step 4: Write to csv
write.csv(flights, "file_name.csv", row.names = FALSE)
干杯。
这是一个 data.table 方法：
library(data.table)
library(lubridate)
dat <- fread("sample_data.txt", col.names = c("dte", "flight", "typ1","typ2","flt_count"))
dat$dte <- as.POSIXct(strptime(dat$dte, format = "%m/%d/%Y %H:%M:%S"), tz = "GMT")
new_dat <- dat[, sum(flt_count), by = list(month(dte),typ1)]
为了方便起见，我重命名了这些列。您也可以根据其他变量将 by 参数编辑为 group/perform 操作。上面代码片段的输出是：
> new_dat
month typ1 V1
1: 1 Arrival 24104
2: 1 Departure 23844
3: 2 Arrival 21365
4: 2 Departure 21394
5: 3 Arrival 24180
6: 3 Departure 24222
这似乎是您要查找的内容。 data.table 对于大型数据集非常有用。
写结果，使用write.csv(new_dat, "new_file.csv", row_names = F)
希望这对您有所帮助。
我相信这要简单得多。请注意，months 的格式与您的略有不同。
library(zoo)
months <- as.yearmon(flights$ReportPeriod, "%m/%d/%Y %H:%M:%S")
agg <- aggregate(FlightOpsCount ~ months + Arrival_Departure, flights, median)
无论如何，我无法得到甚至不接近你问题中预期值的中位数。由于这些是中位数，您的预期似乎有些错误。
如果您想要其他日期格式，可以使用函数 format.Date。
format(as.Date(months), "%Y %B") # or "%B %Y"
有关可能的格式的长列表，请参阅帮助页面 ?strptime。
谢谢大家的帮助！我确实收到了特定月份的正确值，这是我的代码：
#summarize Arrival & Departures through the years
step_1 <- flights %>%
group_by(ReportPeriod, Arrival_Departure) %>%
summarise(sum = sum(FlightOpsCount)) %>%
arrange(ReportPeriod) %>%
ungroup()
#modify date format in ReportPeriod column to receive months
step_2 <- step_1 %>%
transmute(month = months(as.Date(ReportPeriod,"%m/%d/%Y")),
Arrival_Departure,
sum) %>%
group_by(month, Arrival_Departure) %>%
summarise(FlightOpsCount = median(sum)) %>%
write.csv(., "flights_output.csv", row.names = FALSE, quote = FALSE)
但是，我得到的月份是按字母顺序而不是按时间顺序排列的。我在这里的某个地方找到了解决方案，但它不能正常工作，我只得到 NAs。显然我在将任何内容写入 .csv 之前调用它并在 step_2.
的末尾添加 ungroup()step_3 <- step_2 %>%
mutate(month = factor(month.name[month], levels = month.name)) %>%
arrange(month)