在第二個數據幀中基於R列標題標簽的水平值

Question

利用以下示例數據，

Model <- c(1,1,1,1,1,2,2,2,2,4,4,4,4,3,3,3,3,3)
FactorID <- c  ("Factor1",      "Factor2",      "Factor4",      "Factor3",      "Factor5",      "Factor2",      "Factor3",      "Factor4",      "Factor1",      "Factor2",      "Factor3",      "Factor4",      "Factor1",      "Factor3",      "Factor2",      "Factor4",      "Factor5",      "Factor1")

FactorName<- c("SEK", "GBP",  "USD",  "CAD",  "YEN",  "GBP",  "USD",  "CAD",  "EUR",  "CAD",  "EUR",  "USD",  "GBP",  "YEN",  "CAD",  "EUR",  "USD",  "SEK")

a <- data.frame(Model,FactorID,FactorName)

Model <- c(2,1,3,4)
Factor1 <- c(0.054, 0.113,  0.903,  0.720)
Factor2 <- c(0.885, 0.153,  0.708,  0.750)
Factor3 <- c(0.430, 0.989,  0.518,  0.843)
Factor4 <- c(0.533, 0.6328, 0.343,  0.961)
Factor5 <- c("-",     0.055,  0.699,  "-") 

b <- data.frame(Model,Factor1,Factor2,Factor3,Factor4,Factor5)

我希望將數據幀b分成4個數據幀（每個模型一個），並在數據幀a中找到適當的列標題名稱（例如，EUR，GBP等，而不是Factor1，Factor2等）。

Answer 1

你可以做這樣的事情。 它按Model拆分兩個df，然后循環遍歷，將a的列與b的列名匹配，並為新名稱分配適當的幣種。

bSplit <- split(b,b$Model)
aSplit <- split(a,a$Model)

for(i in seq_along(bSplit)){
  names(bSplit[[i]])[-1] <- 
          as.character(aSplit[[i]]$FactorName)[
                match(names(bSplit[[i]])[-1],
                      aSplit[[i]]$FactorID)]
}

bSplit
$`1`
  Model   SEK   GBP   CAD    USD   YEN
2     1 0.113 0.153 0.989 0.6328 0.055

$`2`
  Model   EUR   GBP  USD   CAD NA
1     2 0.054 0.885 0.43 0.533  -

$`3`
  Model   SEK   CAD   YEN   EUR   USD
3     3 0.903 0.708 0.518 0.343 0.699

$`4`
  Model  GBP  CAD   EUR   USD NA
4     4 0.72 0.75 0.843 0.961  -

Answer 2

這是另一種方式。 我首先重排的順序FactorName在a通過轉換FactorID和FactorName字符。 然后，我按Model拆分b 。 在lapply() ，我為每個Model獲取FactorName ，並使用它們來重寫列名稱。

library(dplyr)

arrange(a, Model, FactorID) %>%
mutate_all(funs(as.character(.))) ->a

split(b, f = b$Model) -> whatever

lapply(1:length(ana), function(x){

    foo <- a$FactorName[a$Model == x]
    names(whatever[[x]]) <- c("Model", foo)
    whatever[[x]]

})

#[[1]]
#  Model   SEK   GBP   CAD    USD   YEN
#2     1 0.113 0.153 0.989 0.6328 0.055
#
#[[2]]
#  Model   EUR   GBP  USD   CAD NA
#1     2 0.054 0.885 0.43 0.533  -
#
#[[3]]
#  Model   SEK   CAD   YEN   EUR   USD
#3     3 0.903 0.708 0.518 0.343 0.699
#
#[[4]]
#  Model  GBP  CAD   EUR   USD NA
#4     4 0.72 0.75 0.843 0.961  -

Answer 3

早上好，

我為您提出了一個解決方案，具體方法如下：

library(stringr)
library(plyr)

models = vector("list", length(unique(b$Model)))

for (i in seq_along(1:(ncol(b) - 1))) {
    name <- paste0("model_",i,sep<-"")
    file <- subset(b ,Model == i)

    temp <- a[grep(file[,"Model"],a[,"Model"]) ,]

    for (col in colnames(file)[-1]) {
        colnames(file)[grep(col,colnames(file))] <- as.character(subset(temp,FactorID == col)[1,3])
    }

    models[[i]] = assign(name,file)
}

rm(i,col,temp,file)

models