如何在将丢失的元素保留为空白列的同时跳过循环中的错误？

Question

我有以下数据块：

> dput(data)
structure(c(0.640372781, 0.54596394, 0.364612178, 0.554321638, 
0.623891566, 0.299900389, 0.629781465, 0.502673674, 0.414942748, 
0.485381455, 0.629032253, 0.201974626, 0.549820206, 0.49277897, 
0.299640651, 0.443151949, 0.506297992, 0.259198111, 0.635090505, 
0.597640686, 0.430193856, 0.631067648, 0.662995875, 0.391062922, 
0.632248042, 0.627503454, 0.432827825, 0.418849204, 0.612201188, 
0.227470395, 0.556520484, 0.6095603, 0.414923451, 0.57634896, 
0.543780581, 0.320027087, 0.655818488, 0.648937123, 0.497094053, 
0.429772696, 0.632386262, 0.270060224, 0.564427852, 0.456642259, 
0.492407708, 0.436349654, 0.616355794, 0.248897538, 0.642866477, 
0.555022037, 0.358901689, 0.53184597, 0.606299729, 0.342449093, 
0.667681177, 0.506448197, 0.370292817, 0.555462276, 0.642302168, 
0.42487856, 0.649249462, 0.544035494, 0.394793334, 0.383522657, 
0.557789563, 0.220189788, 0.636151283, 0.547825201, 0.391789202, 
0.653913292, 0.649412792, 0.452257495, 0.648866884, 0.535907987, 
0.392093314, 0.724788138, 0.674157973, 0.494385979, 0.673032345, 
0.450686601, 0.369089571, 0.397124065, 0.502592807, 0.197922003
), class = c("xts", "zoo"), .indexCLASS = "Date", tclass = "Date", .indexTZ = "UTC", tzone = "UTC", index = structure(c(1025049600, 
1025136000, 1025222400, 1025481600, 1025568000, 1025654400), tzone = "UTC", tclass = "Date"), .Dim = c(6L, 
14L), .Dimnames = list(NULL, c("AN8068571086", "BMG3223R1088", 
"BMG4388N1065", "BMG6359F1032", "BMG7496G1033", "BMG812761002", 
"CA88157K1012", "CH0044328745", "CH0048265513", "GB00B4VLR192", 
"GB00B5BT0K07", "GB00B6SLMV12", "GB00BFG3KF26", "GB00BVVBC028"
)))

还有这段代码：

    ######## INPUTS ######

a <- 0.5
b <- 0.6

results <- list() # list containing loop results

#######################


for (i in 1:nrow(data)) {


  input <- as.matrix(data[i,])

  #extract column names with a value between a and b  

  stocks <- matrix(colnames(data[,which(input > a & input < b)]))

  # make a vector with new name for the output
  date   <- head(rownames(input), n=1)

  #rename column  
  colnames(stocks) <- date

  #export to list under "date" name
  results[[date]] <- stocks

}

如果你运行完全照原样你会得到这个错误：

Error in matrix(colnames(data[, which(input > a & input < b)])) : 
  'data' must be of a vector type, was 'NULL'
In addition: Warning messages:
1: In min(j, na.rm = TRUE) :
  no non-missing arguments to min; returning Inf
2: In max(j, na.rm = TRUE) :
  no non-missing arguments to max; returning -Inf

这来自 data 中的第三行，其中不包含介于 0.5 和 0.6 之间的值

matrix(colnames(data[,which(input > a & input < b)]))

在运行上面的代码我运行之后，它将我所有的数据合并在一起并为其他计算做准备：

# merge all results in a list 
max_length <- max(sapply(results ,length))

final_results <- sapply(results, function(x){
  c(x, rep(NA, max_length - length(x)))
})

我需要一种方法来跳过该错误，同时仍将日期名称保留为 final_results 中的空白列。我在想也许是一个 if 函数，这样如果 a 和 b 之间没有值，则使用 colname = date 创建一个空矩阵 (1x1) 应该存储在 results 列表中。

另一个选项是使用 tryCatch 但这将完全省略日期，并且在我这里的数千个之间将不可能找到丢失的列。

Answer 1

如果您使用 tryCatch 函数通过返回具有 NA 值的矩阵来处理您的错误，您将得到一个列，其中包含适当的日期作为名称，并且在 final_results 中仅包含 NA。但是，这将以相同的方式处理所有错误，因此如果您的数据可能引发不同的错误，则可能不是最佳解决方案。

    stocks <- tryCatch(matrix(colnames(data[,which(input > a & input < b)])), 
                       error = function(e) matrix(NA))

Answer 2

快速解决您的任务：

DF <- as.data.frame(data)
DF <- apply(DF, 1, function(x) {
  ifelse(x > a & x < b, x, NA_real_)
})

结果是：

> DF
             2002-06-26 2002-06-27 2002-06-28 2002-07-01 2002-07-02 2002-07-03
AN8068571086         NA  0.5459639         NA  0.5543216         NA         NA
BMG3223R1088         NA  0.5026737         NA         NA         NA         NA
BMG4388N1065  0.5498202         NA         NA         NA  0.5062980         NA
BMG6359F1032         NA  0.5976407         NA         NA         NA         NA
BMG7496G1033         NA         NA         NA         NA         NA         NA
BMG812761002  0.5565205         NA         NA  0.5763490  0.5437806         NA
CA88157K1012         NA         NA         NA         NA         NA         NA
CH0044328745  0.5644279         NA         NA         NA         NA         NA
CH0048265513         NA  0.5550220         NA  0.5318460         NA         NA
GB00B4VLR192         NA  0.5064482         NA  0.5554623         NA         NA
GB00B5BT0K07         NA  0.5440355         NA         NA  0.5577896         NA
GB00B6SLMV12         NA  0.5478252         NA         NA         NA         NA
GB00BFG3KF26         NA  0.5359080         NA         NA         NA         NA
GB00BVVBC028         NA         NA         NA         NA  0.5025928         NA

如果您想删除所有 NA 的列，那么您可以像这样过滤它们：

DF <- DF[
  , apply(DF, 2, function(x) {
  sum(is.na(x)) != length(x)
})]

过滤后的结果为：

> DF
             2002-06-26 2002-06-27 2002-07-01 2002-07-02
AN8068571086         NA  0.5459639  0.5543216         NA
BMG3223R1088         NA  0.5026737         NA         NA
BMG4388N1065  0.5498202         NA         NA  0.5062980
BMG6359F1032         NA  0.5976407         NA         NA
BMG7496G1033         NA         NA         NA         NA
BMG812761002  0.5565205         NA  0.5763490  0.5437806
CA88157K1012         NA         NA         NA         NA
CH0044328745  0.5644279         NA         NA         NA
CH0048265513         NA  0.5550220  0.5318460         NA
GB00B4VLR192         NA  0.5064482  0.5554623         NA
GB00B5BT0K07         NA  0.5440355         NA  0.5577896
GB00B6SLMV12         NA  0.5478252         NA         NA
GB00BFG3KF26         NA  0.5359080         NA         NA
GB00BVVBC028         NA         NA         NA  0.5025928

如何在将丢失的元素保留为空白列的同时跳过循环中的错误？

How to skip an error in loop while preserving the missing element as a blank column?

r

loops

if-statement

try-catch

skip