R - Merge 2 data frames with one column being different

Question

I have 2 data frames of equal column size of 5. The first 4 column names are the same and the last column is different. I report the value (T) in the last column indicating that there is an outlier for each of average & sigma in respective data frames.

My first data frame - df1

    TimeStamp <- c("2015-04-01 11:40:13", "2015-04-03 02:54:45")
    ID <- c("DL1X8", "DL202")
    Avg <- c(38.1517, 0.7131)
    Sig <- c(11.45880000, 0.01257816)
    Outlier_Avg <- c("T","T")
    df1 <- data.frame(TimeStamp, ID, Avg, Sig,Outlier_Avg)


    +---------------------+-------+---------+-------------+-------------+
    |      TimeStamp      |  ID   |   Avg   |     Sig     | Outlier_Avg |
    +---------------------+-------+---------+-------------+-------------+
    | 2015-04-01 11:40:13 | DL1X8 | 38.1517 | 11.45880000 | T           |
    | 2015-04-03 02:54:45 | DL202 | 0.7131  | 0.01257816  | T           |
    +---------------------+-------+---------+-------------+-------------+

My Second data frame - df2

TimeStamp <- c("2015-04-01 11:40:13", "2015-04-04 02:57:45", "2015-04-06 09:54:45")
ID <- c("DL1X8", "DP308","DM3X8")
Avg <- c(38.1517, 24.7131, 0.0234)
Sig <- c(11.4588, 6.0175,0.0665)
Outlier_Sig <- c("T","T","T")
df2 <- data.frame(TimeStamp, ID, Avg, Sig,Outlier_Sig)

+---------------------+-------+---------+---------+-------------+
|      TimeStamp      |  ID   |   Avg   |   Sig   | Outlier_Sig |
+---------------------+-------+---------+---------+-------------+
| 2015-04-01 11:40:13 | DL1X8 | 38.1517 | 11.4588 | T           |
| 2015-04-04 02:57:45 | DP308 | 24.7131 | 6.0175  | T           |
| 2015-04-06 09:54:45 | DM3X8 | 0.0234  | 0.0665  | T           |
+---------------------+-------+---------+---------+-------------+

Desired Output:

I am trying to get a df3 that looks like this

+---------------------+-------+---------+-------------+-------------+-------------+
|      TimeStamp      |  ID   |   Avg   |     Sig     | Outlier_Avg | Outlier_Sig |
+---------------------+-------+---------+-------------+-------------+-------------+
| 2015-04-01 11:40:13 | DL1X8 | 38.1517 | 11.45880000 | T           | T           |
| 2015-04-03 02:54:45 | DL202 | 0.7131  | 0.01257816  | T           | N/A         |
| 2015-04-04 02:57:45 | DP308 | 24.7131 | 6.0175      | N/A         | T           |
| 2015-04-06 09:54:45 | DM3X8 | 0.0234  | 0.0665      | N/A         | T           |
+---------------------+-------+---------+-------------+-------------+-------------+

I tried using merge(df1,df2). It returns only the rows that are matched and hence only 1 row is returned. I need to return all the rows and put N/A as shown above. Could you kindly help me on this?

Thomas Thomas · Accepted Answer · 2015-04-08T14:28:12

Use the all argument:

merge(df1, df2, all = TRUE)
#             TimeStamp    ID     Avg         Sig Outlier_Avg Outlier_Sig
# 1 2015-04-01 11:40:13 DL1X8 38.1517 11.45880000           T           T
# 2 2015-04-03 02:54:45 DL202  0.7131  0.01257816           T        <NA>
# 3 2015-04-04 02:57:45 DP308 24.7131  6.01750000        <NA>           T
# 4 2015-04-06 09:54:45 DM3X8  0.0234  0.06650000        <NA>           T

This is shorthand for using all.x = TRUE and all.y = TRUE, which are separate arguments that let you control which observations from x (df1 in your case) and y (df2 in your case) are included in the merged data.frame. See, for example:

merge(df1, df2, all.x = TRUE)
#             TimeStamp    ID     Avg         Sig Outlier_Avg Outlier_Sig
# 1 2015-04-01 11:40:13 DL1X8 38.1517 11.45880000           T           T
# 2 2015-04-03 02:54:45 DL202  0.7131  0.01257816           T        <NA>

merge(df1, df2, all.y = TRUE)
#             TimeStamp    ID     Avg     Sig Outlier_Avg Outlier_Sig
# 1 2015-04-01 11:40:13 DL1X8 38.1517 11.4588           T           T
# 2 2015-04-04 02:57:45 DP308 24.7131  6.0175        <NA>           T
# 3 2015-04-06 09:54:45 DM3X8  0.0234  0.0665        <NA>           T

R - Merge 2 data frames with one column being different

1 Answers