1

我一直使用 xts 包中的 to.daily 来实现这一点，但是由于升级 R 遇到了该功能的已知错误，所以想找到一种替代方法。

如果可能的话，我想使用 Base R 中的函数。看起来我可以用聚合来做到这一点。我不知道如何开始，所以任何提示将不胜感激。

z <- read.csv("R H2007-EIR_5.csv", sep=",",  header=F)
names(z) <-c("date","time","open","high","low","close","volume")

> head(z,20)
         date  time   open   high    low  close volume
1  13/03/2007 09:10 107.66 107.66 107.66 107.66     10
2  13/03/2007 09:40 107.63 107.63 107.63 107.63     50
3  13/03/2007 09:45 107.64 107.64 107.64 107.64      8
4  13/03/2007 10:00 107.62 107.62 107.62 107.62      7
5  13/03/2007 10:45 107.64 107.65 107.64 107.65     94
6  13/03/2007 11:20 107.77 107.77 107.77 107.77      2
7  14/03/2007 09:00 108.00 108.00 108.00 108.00     10
8  14/03/2007 09:05 108.00 108.01 108.00 108.01     45
9  14/03/2007 11:15 108.05 108.05 108.05 108.05      5
10 14/03/2007 11:25 108.05 108.05 108.05 108.05      1
11 14/03/2007 11:40 108.10 108.10 108.10 108.10     25
12 14/03/2007 12:00 108.10 108.10 108.10 108.10      5
13 14/03/2007 12:30 108.05 108.05 108.05 108.05      5
14 14/03/2007 12:55 108.05 108.05 108.05 108.05    800
15 14/03/2007 13:05 108.02 108.02 108.02 108.02     89
16 14/03/2007 13:30 108.00 108.00 108.00 108.00      5
17 14/03/2007 14:25 107.95 107.95 107.95 107.95      5
18 14/03/2007 15:05 107.95 107.95 107.95 107.95      2
19 14/03/2007 16:00 108.01 108.01 108.01 108.01      6
20 15/03/2007 08:05 107.86 107.90 107.86 107.90      2

> dput(z)
    structure(list(date = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 
3L, 4L, 4L, 4L, 5L, 5L, 5L, 5L, 5L, 6L, 6L, 7L, 7L, 7L, 7L, 7L, 
7L, 7L, 8L, 8L, 8L, 8L, 8L, 8L, 9L, 9L, 9L, 9L, 10L, 10L, 10L, 
10L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 11L, 12L, 12L, 12L, 12L, 
12L, 12L, 12L, 12L, 12L, 12L), .Label = c("13/03/2007", "14/03/2007", 
"15/03/2007", "16/03/2007", "19/03/2007", "20/03/2007", "21/03/2007", 
"22/03/2007", "23/03/2007", "26/03/2007", "27/03/2007", "28/03/2007"
), class = "factor"), time = structure(c(8L, 11L, 12L, 13L, 20L, 
26L, 6L, 7L, 25L, 27L, 28L, 29L, 33L, 34L, 35L, 37L, 42L, 46L, 
54L, 2L, 3L, 24L, 45L, 46L, 11L, 44L, 55L, 11L, 23L, 31L, 32L, 
48L, 10L, 43L, 1L, 14L, 15L, 20L, 31L, 53L, 54L, 36L, 46L, 47L, 
49L, 50L, 51L, 12L, 16L, 30L, 40L, 19L, 31L, 38L, 39L, 4L, 27L, 
31L, 33L, 41L, 43L, 50L, 52L, 5L, 7L, 9L, 12L, 17L, 18L, 20L, 
21L, 22L, 23L), .Label = c("08:00", "08:05", "08:10", "08:30", 
"08:55", "09:00", "09:05", "09:10", "09:15", "09:30", "09:40", 
"09:45", "10:00", "10:05", "10:15", "10:25", "10:30", "10:35", 
"10:40", "10:45", "10:50", "10:55", "11:00", "11:10", "11:15", 
"11:20", "11:25", "11:40", "12:00", "12:05", "12:15", "12:20", 
"12:30", "12:55", "13:05", "13:25", "13:30", "13:35", "13:55", 
"14:00", "14:15", "14:25", "14:30", "14:35", "14:45", "15:05", 
"15:10", "15:15", "15:20", "15:25", "15:30", "15:40", "15:55", 
"16:00", "16:10"), class = "factor"), open = c(107.66, 107.63, 
107.64, 107.62, 107.64, 107.77, 108, 108, 108.05, 108.05, 108.1, 
108.1, 108.05, 108.05, 108.02, 108, 107.95, 107.95, 108.01, 107.86, 
107.86, 107.88, 107.8, 107.79, 107.78, 107.64, 107.75, 107.61, 
107.6, 107.55, 107.55, 107.48, 107.4, 107.43, 107.39, 107.48, 
107.53, 107.56, 107.47, 107.39, 107.39, 107.37, 107.3, 107.3, 
107.29, 107.29, 107.29, 107.16, 107.13, 107.11, 107.06, 106.8, 
106.72, 106.7, 106.72, 106.62, 106.72, 106.58, 106.58, 106.62, 
106.66, 106.73, 106.67, 106.61, 106.6, 106.6, 106.57, 106.61, 
106.6, 106.59, 106.6, 106.62, 106.65), high = c(107.66, 107.63, 
107.64, 107.62, 107.65, 107.77, 108, 108.01, 108.05, 108.05, 
108.1, 108.1, 108.05, 108.05, 108.02, 108, 107.95, 107.95, 108.01, 
107.9, 107.86, 107.89, 107.8, 107.79, 107.78, 107.64, 107.75, 
107.61, 107.6, 107.55, 107.55, 107.48, 107.4, 107.43, 107.39, 
107.48, 107.53, 107.56, 107.47, 107.39, 107.39, 107.37, 107.3, 
107.3, 107.29, 107.29, 107.29, 107.16, 107.13, 107.11, 107.06, 
106.8, 106.72, 106.7, 106.72, 106.62, 106.72, 106.58, 106.58, 
106.62, 106.66, 106.73, 106.67, 106.62, 106.6, 106.6, 106.57, 
106.61, 106.6, 106.63, 106.6, 106.65, 106.65), low = c(107.66, 
107.63, 107.64, 107.62, 107.64, 107.77, 108, 108, 108.05, 108.05, 
108.1, 108.1, 108.05, 108.05, 108.02, 108, 107.95, 107.95, 108.01, 
107.86, 107.86, 107.88, 107.8, 107.79, 107.78, 107.64, 107.75, 
107.61, 107.6, 107.55, 107.55, 107.48, 107.29, 107.43, 107.39, 
107.48, 107.53, 107.56, 107.47, 107.39, 107.39, 107.37, 107.3, 
107.3, 107.29, 107.29, 107.29, 107.16, 107.13, 107.11, 107.06, 
106.8, 106.72, 106.7, 106.72, 106.62, 106.72, 106.58, 106.58, 
106.62, 106.66, 106.73, 106.67, 106.61, 106.6, 106.6, 106.57, 
106.61, 106.6, 106.59, 106.6, 106.61, 106.65), close = c(107.66, 
107.63, 107.64, 107.62, 107.65, 107.77, 108, 108.01, 108.05, 
108.05, 108.1, 108.1, 108.05, 108.05, 108.02, 108, 107.95, 107.95, 
108.01, 107.9, 107.86, 107.89, 107.8, 107.79, 107.78, 107.64, 
107.75, 107.61, 107.6, 107.55, 107.55, 107.48, 107.3, 107.43, 
107.39, 107.48, 107.53, 107.56, 107.47, 107.39, 107.39, 107.37, 
107.3, 107.3, 107.29, 107.29, 107.29, 107.16, 107.13, 107.11, 
107.06, 106.8, 106.72, 106.7, 106.72, 106.62, 106.72, 106.58, 
106.58, 106.62, 106.66, 106.73, 106.67, 106.61, 106.6, 106.6, 
106.57, 106.61, 106.6, 106.61, 106.6, 106.65, 106.65), volume = c(10L, 
50L, 8L, 7L, 94L, 2L, 10L, 45L, 5L, 1L, 25L, 5L, 5L, 800L, 89L, 
5L, 5L, 2L, 6L, 2L, 4L, 178L, 5L, 5L, 10L, 1L, 1L, 1L, 2L, 50L, 
44L, 100L, 400L, 91L, 1L, 100L, 100L, 3L, 1L, 79L, 21L, 28L, 
80L, 20L, 20L, 31L, 49L, 5L, 1L, 25L, 1L, 20L, 284L, 2368L, 454L, 
18L, 43L, 11L, 547L, 18L, 1L, 3L, 253L, 200L, 5L, 35L, 30L, 50L, 
50L, 172L, 99L, 1728L, 82L)), .Names = c("date", "time", "open", 
"high", "low", "close", "volume"), class = "data.frame", row.names = c(NA, 
-73L))

期望的输出：

         date   open   high    low  close
 1 13/03/2007 107.66 107.77 107.62 107.77
 2 14/03/2007 108.00 108.10 107.95 108.01
 3 15/03/2007 107.86 107.90 107.86 107.90
4

3 回答 3

1

是这样的吗？

o <- aggregate(data = z, open ~ date, head, 1)
o$max <- aggregate(data = z, high ~ date, max)$high
o$min <- aggregate(data = z, low ~ date, min)$low
o$close <- aggregate(data = z, close ~ date, tail, 1)$close

#         date   open    max    min  close
# 1 13/03/2007 107.66 107.77 107.62 107.77
# 2 14/03/2007 108.00 108.10 107.95 108.01
# 3 15/03/2007 107.86 107.90 107.86 107.90
于 2013-03-07T08:57:46.697 回答
1

这样的事情应该可以解决问题：

todaily <- function(z){
    zperiod<-split(z,cut(strptime(paste(z$date, z$time), "%d/%m/%Y %H:%M"), "day"))
    zperiod<-zperiod[sapply(zperiod,nrow)!=0]
    res<-do.call(rbind,lapply(zperiod, 
                  function(x)c(x$open[1],
                               max(c(x$open,x$max, x$min,x$close), na.rm=TRUE),     
                               min(c(x$open,x$max, x$min,x$close), na.rm=TRUE), 
                               x$close[nrow(x)])))
    colnames(res)<-c("open","max","min","close")
    res
    }
todaily(z)
             open    max    min  close
2007-03-13 107.66 107.77 107.62 107.77
2007-03-14 108.00 108.10 107.95 108.01
2007-03-15 107.86 107.90 107.86 107.90

当然，您可以在可以处理的任何时间段内对其进行修改cut.POSIXt（请参阅?cut.POSIXt）：

toperiod <- function(z, period="day"){
    zperiod<-split(z,cut(strptime(paste(z$date, z$time), "%d/%m/%Y %H:%M"), period))
    zperiod<-zperiod[sapply(zperiod,nrow)!=0]
    res<-do.call(rbind,lapply(zperiod, 
                  function(x)c(x$open[1],
                               max(c(x$open,x$max, x$min,x$close), na.rm=TRUE),     
                               min(c(x$open,x$max, x$min,x$close), na.rm=TRUE), 
                               x$close[nrow(x)])))
    colnames(res)<-c("open","max","min","close")
    res
    }

使用您上传的完整数据集，这是最后一个函数可以执行的操作：

toperiod(z, "weeks")
                      open    max    min  close
2007-03-12 00:00:00 107.66 108.10 107.62 107.75
2007-03-19 00:00:00 107.61 107.61 107.06 107.06
2007-03-26 01:00:00 106.80 106.80 106.57 106.65

toperiod(z, "2 weeks")
                      open   max    min  close
2007-03-12 00:00:00 107.66 108.1 107.06 107.06
2007-03-26 01:00:00 106.80 106.8 106.57 106.65
于 2013-03-07T08:58:11.143 回答
0

这是一个混合xts和base功能的解决方案。

首先，我创建我的 xts 对象。

INDEX <- strptime(paste(dat[,1],dat[,2],sep=' '),'%d/%m/%Y %H:%M')
dat.xts <- xts(dat[,-c(1,2)],INDEX)

最终解决方案是此 xts 对象的子集。我endpoints用来获取最终数据的子集，但使用了错误的核心数据。

dat.xts[INDEX,-c(5)]
                      open   high    low  close
2007-03-13 11:20:00 107.77 107.77 107.77 107.77
2007-03-14 16:00:00 108.01 108.01 108.01 108.01
2007-03-15 08:05:00 107.86 107.90 107.86 107.90

现在我计算核心数据，使用lapply

xx <- lapply(1:(length(INDEX) - 1), function(y) {
  xi <- as.data.frame(dat.xts[(INDEX[y] + 1):INDEX[y + 1]])
    res <- c(
      xi$open[1],
      max(c(xi$open,x$max, xi$min,xi$close), na.rm=TRUE),     
      min(c(xi$open,x$max, xi$min,xi$close), na.rm=TRUE), 
      x$close[nrow(xi)])
  })

coredata(res) <- do.call(rbind,xx)

要获得所需的结果：

                       open   high    low  close
2007-03-13 11:20:00 107.66 107.77 107.62 107.77
2007-03-14 16:00:00 108.00 108.10 107.95 108.05
2007-03-15 08:05:00 107.86 107.90 107.86 107.66
于 2013-03-07T11:30:10.113 回答