我正在使用eventstudies package。我正在使用phys2eventtime(..)来设置我的数据。但是我得到了
Error in `colnames<-`(`*tmp*`, value = integer(0)) :
attempt to set 'colnames' on an object with less than two dimensions
我的猜测是，我的数据格式错误。当我查看使用示例中的函数的示例数据时：
> es <- phys2eventtime(z=StockPriceReturns, events=SplitDates, width=10)
> (str(StockPriceReturns))
'zoo' series from 2000-04-03 to 2013-03-28
Data: num [1:3246, 1:30] NA NA NA NA NA NA NA NA NA NA ...
- attr(*, "dimnames")=List of 2
..$ : NULL
..$ : chr [1:30] "Bajaj.Auto" "BHEL" "Bharti.Airtel" "Cipla" ...
Index: Date[1:3246], format: "2000-04-03" "2000-04-04" "2000-04-05" "2000-04-06" ...
NULL
与此相比，我的数据看起来像是：
> (str(zoo_Data))
'zoo' series from 2002-01-01 to 2013-08-20
Data: num [1:3036] 183 183 186 191 191 ...
Index: Date[1:3036], format: "2002-01-01" "2002-01-02" "2002-01-03" "2002-01-04" ...
NULL
这是不同的：
- attr(*, "dimnames")=List of 2 查看原始数据时也可以看到这一点：
示例数据：
> (head(StockPriceReturns))
Bajaj.Auto BHEL Bharti.Airtel Cipla Coal.India Dr.Reddy
2000-04-03 NA 4.9171044 NA 6.810041 NA -3.2541653
2000-04-04 NA -8.3348496 NA -3.368606 NA -8.3353739
2000-04-05 NA 0.3305788 NA 0.836825 NA 0.2616345
2000-04-06 NA -2.7605266 NA -2.466056 NA -1.8941289
2000-04-07 NA 3.2543548 NA 7.690426 NA 7.6961041
2000-04-10 NA 3.3107586 NA 6.154276 NA 6.4769648
我的数据：
> (head(zoo_Data))
2002-01-01 2002-01-02 2002-01-03 2002-01-04 2002-01-07 2002-01-08
182.83 182.83 186.40 190.57 191.17 193.25
df data：
我正在使用我的数据构建zoo对象：
> dfToZoo <- function(df) {
require(zoo)
date <- as.Date(df[, 1], format = '%d.%m.%Y')
#TODO have a look if the column are rightly named
with(df, zoo(TotalReturns, date))
}
csv_data <- read.csv(..., header = TRUE, sep = ";",stringsAsFactors=FALSE)
totalReturns <- (as.double(gsub(",",".",csv_data$TotalReturn)))
df <- data.frame(csv_data$Date, totalReturns)
names(df) <- c("Date" ,"TotalReturns")
zoo_Data <- dfToZoo(df)
如何将attr(*, "dimnames")=List of 2添加到我的数据中？
感谢您的回复！
更新1
这是libary包中的example data code：
library(eventstudies)
library(zoo)
###########################
# Load data
(data(SplitDates))
(data(StockPriceReturns))
data <- StockPriceReturns
(head(StockPriceReturns))
es <- phys2eventtime(z=StockPriceReturns, events=SplitDates, width=10)
es.w <- window(es$z.e, start=-10, end=10)
SplitDates[1:3,]
更新2
这就是我自己的数据格式化方式：
> (str(s_dates))
'data.frame': 36799 obs. of 2 variables:
$ unit: chr "ZAE000006284" "ZAE000006284" "ZAE000006284" "ZAE000006284" ...
$ when: Date, format: "2003-12-18" "2005-04-06" ...
NULL
> (str(zoo_Data))
'zoo' series from 2002-01-01 to 2013-08-20
Data: num [1:3036] 183 183 186 191 191 ...
Index: Date[1:3036], format: "2002-01-01" "2002-01-02" "2002-01-03" "2002-01-04" ...
NULL
这是事件研究包中数据的格式：
> (str(SplitDates))
'data.frame': 22 obs. of 2 variables:
$ unit: chr "BHEL" "Bharti.Airtel" "Cipla" "Coal.India" ...
$ when: Date, format: "2011-10-03" "2009-07-24" ...
NULL
> (str(StockPriceReturns))
'zoo' series from 2000-04-03 to 2013-03-28
Data: num [1:3246, 1:30] NA NA NA NA NA NA NA NA NA NA ...
- attr(*, "dimnames")=List of 2
..$ : NULL
..$ : chr [1:30] "Bajaj.Auto" "BHEL" "Bharti.Airtel" "Cipla" ...
Index: Date[1:3246], format: "2000-04-03" "2000-04-04" "2000-04-05" "2000-04-06" ...
NULL
更新3
可再现的示例
我最初通过不同的csv文件读取数据。请参阅我的可重复示例：
library(eventstudies)
library(zoo)
s_dates <-
structure(list(unit = c("ZAE000006284", "ZAE000006284", "ZAE000006284",
"ZAE000006284", "ZAE000006284", "XS0430907989", "XS0430907989"
), when = structure(c(12404, 12879, 12879, 12880, 12930, 14411,
14411), class = "Date")), .Names = c("unit", "when"), row.names = c(NA,
7L), class = "data.frame")
zoo_Data <-
structure(c(182.83, 182.83, 186.4, 190.57, 191.17, 193.25, 190.57
), index = structure(c(11688, 11689, 11690, 11691, 11694, 11695,
11696), class = "Date"), class = "zoo")
es <- phys2eventtime(z=zoo_Data, events=s_dates, width=10)
Error in `colnames<-`(`*tmp*`, value = integer(0)) :
attempt to set 'colnames' on an object with less than two dimensions
> es.w <- window(es$z.e, start=-10, end=10)
Error in window(es$z.e, start = -10, end = 10) : object 'es' not found
更新4
好的，当我尝试转换zoo_data：
> library(eventstudies)
> library(zoo)
>
> s_dates <- dput(head(s_dates,30))
structure(list(unit = c("ZAE000006284", "ZAE000006284", "ZAE000006284",
"ZAE000006284", "ZAE000006284", "XS0430907989", "XS0430907989",
"XS0302626899", "XS0302626899", "XS0302626899", "XS0302626899",
"XS0302626899", "XS0302626899", "XS0266838746", "XS0187043079",
"XS0187043079", "XS0187043079", "XF0000TZ7757", "XF0000AK5197",
"XF0000AK5197", "XF0000AK5197", "XF0000AK5197", "XF0000AK5197",
"USU02681027", "USU026281027", "USU026281027", "USU026281027",
"USU026281027", "USU026281027", "USU026281027"), when = structure(c(12404,
12879, 12879, 12880, 12930, 14411, 14411, 14599, 14600, 15134,
15139, 15328, 15328, 13913, 14330, 14335, 14593, 13049, 12953,
12954, 12954, 12954, 12955, 12934, 13537, 13537, 13537, 13648,
13649, 13649), class = "Date")), .Names = c("unit", "when"), row.names = c(NA,
30L), class = "data.frame")
> zoo_Data <- dput(head(zoo_Data,30))
structure(c(182.83, 182.83, 186.4, 190.57, 191.17, 193.25, 190.57,
184.02, 181.34, 172.11, 169.73, 160.2, 175.09, 172.11, 170.92,
176.58, 171.51, 170.92, 173.9, 168.54, 167.34, 166.75, 166.45,
167.34, 164.37, 159.01, 158.11, 154.84, 156.63, 161.99), index = structure(c(11688,
11689, 11690, 11691, 11694, 11695, 11696, 11697, 11698, 11701,
11702, 11703, 11704, 11705, 11708, 11709, 11710, 11711, 11712,
11715, 11716, 11717, 11718, 11719, 11722, 11723, 11724, 11725,
11726, 11729), class = "Date"), class = "zoo")
>
> n = 20 ## number of observation
> cn <- unique(s_dates$unit) ## get response variable
>
> as.xts(zoo_Data)
>
> names(zoo_Data) <- cn
>
> es <- phys2eventtime(z=zoo_Data), events=s_dates, width=10)
Error in `colnames<-`(`*tmp*`, value = integer(0)) :
attempt to set 'colnames' on an object with less than two dimensions
> es.w <- window(es$z.e, start=-3, end=3)
Error in window(es$z.e, start = -3, end = 3) : object 'es' not found
答案 0 :(得分：5)
phys2eventtime将实际日期转换为事件日期。
因此，鉴于动物园时间序列和事件日期，它将物理日期转换为事件日期。这意味着，如果它在动物园对象的范围内找到事件日期，则此日期将变为零，并且所有其他日期也会相应地移动。
所以给这个事件日期对象：
unit when
1 ZAE000006284 2003-12-18
2 ZAE000006284 2005-04-06
3 ZAE000006284 2005-04-06
4 ZAE000006284 2005-04-07
5 ZAE000006284 2005-05-27
6 XS0430907989 2009-06-16
7 XS0430907989 2009-06-16
您的物理时间序列日期（动物园对象）应满足2个条件：
现在我将创建满足以下两个条件的时间序列：多时间序列（xts对象），其索引以s_dates的范围下降并且列名在s_dates单位值内。
n = 20 ## number of observation
cn <- unique(s_dates$unit) ## get response variable
rr = range(s_dates$when) ## get date range
## COND2
index = seq(rr[1],rr[2],length.out=n)
mat = matrix(round(runif(n*2,min=150,max=200),2),ncol=2)
ev.dat = xts(mat,index)
## COND1
names(ev.dat) <- cn
检查我们的时间系列对象：
head(ev.dat)
ZAE000006284 XS0430907989
2003-12-18 183.27 152.89
2004-04-01 195.23 172.78
2004-07-16 190.18 164.92
2004-10-29 182.24 191.19
2005-02-12 151.78 195.81
2005-05-29 153.78 189.27
毫不奇怪，它运作良好：
(es <- phys2eventtime(z=ev.dat, events=s_dates, width=5))
(es.w <- window(es$z.e, start=-10, end=10))
head(es.w)
2 3 4 5
-4 197.28 197.28 197.28 197.28
-3 158.54 158.54 158.54 158.54
-2 180.56 180.56 180.56 180.56
-1 194.41 194.41 194.41 194.41
0 157.64 157.64 157.64 157.64
1 151.06 151.06 151.06 151.06
作为奖励，我创建了一个小功能2条件：
check.phys2ev <- function(ev.dat,s_dates){
is.sucess = lapply(s_dates$when,function(when){
inter = findInterval(when, index(ev.dat))
inter < nrow(ev.dat) && inter >1
})
sum(unlist(is.sucess)) >0 &
all(colnames(ev.dat) %in% s_dates$unit)
}
check.phys2ev(ev.dat,s_dates)
[1] TRUE
答案 1 :(得分：2)
@agstudy已经提出了一个非常好的答案，以多变量时间序列为例。因为您似乎有一个单变量的时间序列，所以我希望在创建zoo对象时添加一些关于如何避免强制使用“少于两维的对象”的内容。这是一个可能的解决方案，我在时间序列中添加一个虚拟列，并在使用drop = FALSE时选择焦点列。
从数据框中的单变量时间序列开始。由于zoo对象和events中的日期（由@agstudy描述）不匹配，以及zoo对象的事实，第一次尝试显然不起作用不到两个维度。
df <- data.frame(date = index(zoo_Data),
ZAE000006284 = coredata(zoo_Data))
df
# convert to zoo object
z1 <- zoo(x = df[ , -1], order.by = df$date)
z1
phys2eventtime(z = z, events = s_dates, width = 10)
# Error in `colnames<-`(`*tmp*`, value = integer(0)) :
# attempt to set 'colnames' on an object with less than two dimensions
在s_dates中添加一个与最大日期一样大的日期，这里仅用作快速而肮脏的方式来实现“COND2”（sensu @agstudy）。
df <- data.frame(date = c(index(zoo_Data), max(s_dates$when)),
ZAE000006284 = c(coredata(zoo_Data), 200))
df
z1 <- zoo(x = df[ , -1], order.by = df$date)
z1
phys2eventtime(z = z1, events = s_dates, width = 10)
# Error in `colnames<-`(`*tmp*`, value = integer(0)) :
# attempt to set 'colnames' on an object with less than two dimensions
避免强制传播
在单变量时间序列中添加虚拟列。选择焦点列后，使用drop = FALSE以避免强制转换为矢量。然后将单变量（但是二维）时间序列转换为zoo：
df <- data.frame(date = c(index(zoo_Data), max(s_dates$when)),
ZAE000006284 = c(coredata(zoo_Data), 200),
ZAE000006284_2 = c(coredata(zoo_Data), 200))
df
z2 <- zoo(x = df[ , "ZAE000006284", drop = FALSE], order.by = df$date)
z2
str(z2)
....
attr(*, "dimnames")=List of 2
phys2eventtime(z = z2, events = s_dates, width = 10)
# seems to work
答案 2 :(得分：1)
感谢大家指出这一点并进行了精彩的讨论。作为该软件包的作者之一，我看到了问题所在。如果您查看phys2eventtime函数，则需要一个至少包含一列的zoo系列，如上面的答案中所述。除非zoo提供给下标函数，否则data.frame / drop = FALSE会删除最后一列并将其转换为向量。
正如@Henrik和@agstudy所建议的那样，我们无法在drop = FALSE函数中真正执行phys2eventtime，因为函数不知道系列名称。它依赖于系列名称对象的colnames。对于单变量系列，如果对象是向量，则无法知道系列名称并将其与事件对象匹配。
我现在看到的唯一方法是创建一列zoo个对象。建议当然是受欢迎的。
if (is.null(ncol(z))) {
stop(paste(deparse("z"), "should be a zoo series with at least one column."))
}
代码可在此处找到：https://r-forge.r-project.org/scm/viewvc.php/?root=eventstudies
请加入包讨论邮件列表：http://lists.r-forge.r-project.org/pipermail/eventstudies-discussion/