如何在具有两列共同的两个不同数据集中查找单个值（基于条件）

Question

我有两个不同的数据集，如下所示：

当且仅当：两个数据集中的 ID 和 Date 列都匹配时，我需要在数据集 #2 中添加 call volume、put volume 和 total volume 列。我根据数据集 #1 中第 3 列中的值（C 代表调用，P 代表看跌，T 代表总计）来区分看涨、看跌和总计。

我正在运行此代码，但它不起作用（仅显示调用示例，同样的规则适用于 put 和 totals）。

dataset2$call_volume <- if(dataset1$optiontype== "C")
{ dataset1$volume [ match (
                          interaction(dataset2$ID,dataset2$date),
                          interaction(dataset1$ID,dataset1$date)
                                         )]}

有没有人建议我如何继续编写代码？非常感谢！

> dput(dataset1)
structure(list(ID = c(44652, 44652, 44652, 56266, 56266, 56266, 
44652, 44652, 44652, 56266, 56266, 56266), date = c("1997/01/02", 
"1997/01/02", "1997/01/02", "1997/01/02", "1997/01/02", "1997/01/02", 
"1997/01/03", "1997/01/03", "1997/01/03", "1997/01/03", "1997/01/03", 
"1997/01/03"), `option type (C,P,T: for calls, puts, and total)` = c("C", 
"P", "T", "C", "P", "T", "C", "P", "T", "C", "P", "T"), volume = c(34, 
250, 284, 30, 0, 30, 1443, 211, 1654, 4490, 826, 5316)), row.names = c(NA, 
-12L), class = c("tbl_df", "tbl", "data.frame"))

> dput(dataset2)
structure(list(ID = c(44652, 44652, 44652, 56266, 56266, 56266
), date = c("1997/01/02", "1997/01/03", "1997/01/04", "1997/01/02", 
"1997/01/03", "1997/01/04"), `call volume` = c(NA, NA, NA, NA, 
NA, NA), `put volume` = c(NA, NA, NA, NA, NA, NA), `total volume` = c(NA, 
NA, NA, NA, NA, NA)), row.names = c(NA, -6L), class = c("tbl_df", 
"tbl", "data.frame"))

更新：我在两个数据集中还有许多其他列彼此完全不同，唯一常见的是下面图片和数据集中显示的列。

Answer 1

我认为这是一个 x/y 问题。我认为您实际上是在尝试将dataset1 转换为宽格式以填充dataset2。之后你可以left_join这两个框架。

library(tidyr)
library(dplyr)

names(dataset1)[3] <- "option_type"

dataset2 %>% 
  dplyr::select(-`call volume`, -`put volume`, -`total volume`) %>%
  left_join(dataset1 %>% 
  tidyr::pivot_wider(names_from = "option_type", values_from = "volume") %>%
  rename("Call Volume" = C, "Put Volume" = P, "Total Volume" = `T`),
  by = c("ID", "date"))
#> # A tibble: 6 x 5
#>      ID date       `Call Volume` `Put Volume` `Total Volume`
#>   <dbl> <chr>              <dbl>        <dbl>          <dbl>
#> 1 44652 1997/01/02            34          250            284
#> 2 44652 1997/01/03          1443          211           1654
#> 3 44652 1997/01/04            NA           NA             NA
#> 4 56266 1997/01/02            30            0             30
#> 5 56266 1997/01/03          4490          826           5316
#> 6 56266 1997/01/04            NA           NA             NA

^{由reprex package (v0.3.0) 于 2020 年 10 月 7 日创建}

Answer 2

如果我理解正确，如果匹配，您希望 dataset2 具有来自 dataset1 的值，如果不匹配，则需要 NA。

如果是这种情况，您需要使用left_join。

如果没有，请使用所需的输出更新您的问题。

library(tidyverse)

d1 <- structure(list(ID = c(44652, 44652, 44652, 56266, 56266, 56266, 
                            44652, 44652, 44652, 56266, 56266, 56266), date = c("1997/01/02", 
                                                                                "1997/01/02", "1997/01/02", "1997/01/02", "1997/01/02", "1997/01/02", 
                                                                                "1997/01/03", "1997/01/03", "1997/01/03", "1997/01/03", "1997/01/03", 
                                                                                "1997/01/03"), `option type (C,P,T: for calls, puts, and total)` = c("C", 
                                                                                                                                                     "P", "T", "C", "P", "T", "C", "P", "T", "C", "P", "T"), volume = c(34, 
                                                                                                                                                                                                                        250, 284, 30, 0, 30, 1443, 211, 1654, 4490, 826, 5316)), row.names = c(NA, 
                                                                                                                                                                                                                                                                                               -12L), class = c("tbl_df", "tbl", "data.frame"))


d2 <- structure(list(ID = c(44652, 44652, 44652, 56266, 56266, 56266
), date = c("1997/01/02", "1997/01/03", "1997/01/04", "1997/01/02", 
            "1997/01/03", "1997/01/04"), `call volume` = c(NA, NA, NA, NA, 
                                                           NA, NA), `put volume` = c(NA, NA, NA, NA, NA, NA), `total volume` = c(NA, 
                                                                                                                                 NA, NA, NA, NA, NA)), row.names = c(NA, -6L), class = c("tbl_df", 
                                                                                                                                                                                         "tbl", "data.frame"))

d1_longer <- d1 %>%
  pivot_wider(names_from = `option type (C,P,T: for calls, puts, and total)`, values_from = volume) %>%
  rename(`call volume` = `C`, `put volume` = `P`, `total volume` = `T`)

d2 %>%
  select(ID, date) %>%
  left_join(d1_longer)
#> Joining, by = c("ID", "date")
#> # A tibble: 6 x 5
#>      ID date       `call volume` `put volume` `total volume`
#>   <dbl> <chr>              <dbl>        <dbl>          <dbl>
#> 1 44652 1997/01/02            34          250            284
#> 2 44652 1997/01/03          1443          211           1654
#> 3 44652 1997/01/04            NA           NA             NA
#> 4 56266 1997/01/02            30            0             30
#> 5 56266 1997/01/03          4490          826           5316
#> 6 56266 1997/01/04            NA           NA             NA

^{由reprex package (v0.3.0) 于 2020 年 10 月 7 日创建}