R：使用另一个数据框中的列名，条件和值在一个数据框中创建一个新列

Question

Consider a base data frame as:

data <-  data.frame(amount_bin = c("10K-25K", "25K-35K", "35K-45K", "45K-50K", "50K+", "10K-25K", "25K-35K", "35K-45K", "45K-50K", "50K+", "10K-25K", "25K-35K", "35K-45K", "45K-50K", "50K+"),
                   risk_score = c("0-700", "700-750", "750-800", "800-850", "850-900", "0-700", "700-750", "750-800", "800-850", "850-900", "0-700", "700-750", "750-800", "800-850", "850-900"))

and grouping information in another data frame as:

group_info <- data.frame(variable = c("amount_bin_group", "amount_bin_group", "amount_bin_group", "amount_bin_group", "amount_bin_group",
                                 "risk_score_group", "risk_score_group", "risk_score_group", "risk_score_group", "risk_score_group"),
                    bin = c("10K-25K", "25K-35K", "35K-45K", "45K-50K", "50K+",
                            "0-700", "700-750", "750-800", "800-850", "850-900"),
                    group = c("1", "1", "2", "2", "3",
                              "a", "a", "a", "b", "b"))

I want to make 2 columns in base data frame (data) called "amount_bin_group" and "risk_score_group", which takes values from the column group_info$group when bin columns from group_info and data are the same. For simplicity, let's assume that the base column will always be the group_info$variable name minus the "group" string. That implies, when we want to create column amount_bin_group, base column will always be amount_bin in the base data frame.

The expected result data frame is:

final_data <-  data.frame(amount_bin = c("10K-25K", "25K-35K", "35K-45K", "45K-50K", "50K+", "10K-25K", "25K-35K", "35K-45K", "45K-50K", "50K+", "10K-25K", "25K-35K", "35K-45K", "45K-50K", "50K+"),
                   risk_score = c("0-700", "700-750", "750-800", "800-850", "850-900", "0-700", "700-750", "750-800", "800-850", "850-900", "0-700", "700-750", "750-800", "800-850", "850-900"),
                   amount_bin_group = c("1", "1", "2", "2", "3", "1", "1", "2", "2", "3", "1", "1", "2", "2", "3"),
                   risk_score_group = c("a", "a", "a", "b", "b", "a", "a", "a", "b", "b", "a", "a", "a", "b", "b"))

A solution that I just thought is iteratively merge the data frames ie :

final_data <- merge(data, group_info[, c("bin", "group")], by.x = "amount_bin", by.y = "bin")

final_data$amount_bin_group <- final_data$group
final_data$group <- NULL

But, I am sure there can be a more efficient solution. Please note that there are multiple such columns and not just two. So, maybe a loop would help.

Answer 1

Your group_info is just way over-tidy. I can't believe I'm actually saying that. By breaking that into either two dataframes, or breaking each half into it's own column, you enable yourself to do a simple left join to get the answer.

final_data_calc <- data %>%
  left_join(
    group_info %>% 
      filter(variable == 'amount_bin_group') %>% 
      rename(amount_bin_group = group,amount_bin = bin) %>% 
      select(-variable)
  ) %>%
  left_join(
    group_info %>% 
      filter(variable == 'risk_score_group') %>% 
      rename(risk_score_group = group,risk_score = bin) %>% 
      select(-variable)
  )

#   amount_bin risk_score amount_bin_group risk_score_group
#1     10K-25K      0-700                1                a
#2     25K-35K    700-750                1                a
#3     35K-45K    750-800                2                a
#4     45K-50K    800-850                2                b
#5        50K+    850-900                3                b
#6     10K-25K      0-700                1                a
#7     25K-35K    700-750                1                a
#8     35K-45K    750-800                2                a
#9     45K-50K    800-850                2                b
#10       50K+    850-900                3                b
#11    10K-25K      0-700                1                a
#12    25K-35K    700-750                1                a
#13    35K-45K    750-800                2                a
#14    45K-50K    800-850                2                b
#15       50K+    850-900                3                b

Answer 2

You could just use a for loop to keep merging on the different sets:

for (i in unique(group_info$variable)) {
  data <- merge(
    data, group_info[group_info$variable==i,c("bin","group")],
    by.x=sub("_group","",i), by.y="bin"
  )
  names(data)[names(data)=="group"] <- i
}