I would like to group the rows of this dataset by Index and then sum the rows by common index

问题

I would like to group the rows of this dataset by MemberID.

This is a snipet of my dataset "Claims":

  MemberID    SopLos    DIH
1     54         0       1
2     55         1       2
3     56         2       3  
4     67         0       5
5     55         1       1
6     54         0       1
7     55         1       2
8     56         2       3  
9     67         0       5
10    55         1       1

My desired data frame:

  MemberID    SopLos    DIH
1     54         0       1
2     54         0       1
3     55         1       1
4     55         1       2
5     55         1       1
6     55         1       2
7     56         2       3  
8     56         2       3  
9     67         0       5
10    67         0       5

Then I would like to sum the rows by common ID's resulting in the following data frame

 MemberID    SopLos    DIH
1     54         0       2
2     55         4       6  
3     56         4       6  
4     67         0       10

If you could give me any help I would greatly appreciate it.

回答1:

Assuming your data is in a dataframe called dat:

require(plyr)
ddply(dat,.(MemberID),summarise,SopLos = sum(SopLos),DIH = sum(DIH))
  MemberID SopLos DIH
1       54      0   2
2       55      4   6
3       56      4   6
4       67      0  10

回答2:

Since I only have base functions available at the moment, here's another solution. Assuming your data is in a dataframe called df:

aggregate(df[c("SopLos","DIH")],list(MemberId = df$MemberID),sum)

  MemberId SopLos DIH
1       54      0   2
2       55      4   6
3       56      4   6
4       67      0  10

If you have a lot of variables to sum or a dataset that changes frequently and don't you wish to type out all the names, you could also try:

aggregate(df[-(grep("^MemberID$",names(df)))],list(MemberId = df$MemberID),sum)

来源：https://stackoverflow.com/questions/9593056/i-would-like-to-group-the-rows-of-this-dataset-by-index-and-then-sum-the-rows-by