aggregate on more than one variable at a time using loop and/or apply in R

Question

I'm working with programming language R on a dataframe ( data ) that look like this:

   ID     t    P1    P2    P3    P4
    <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
 1 100003     0     5     4     3     2
 2 100003     0     6     2     1     3
 3 100013     0     6     5     7     3
 4 100013     0     4     5     4     1
 5 100014     0     1     1     1     1
 6 100014     0     1     1     1     1
 7 100015     0     6     6     1     1
 8 100015     0     6     6     1     1
 9 100044     0     6     2     5     1
10 100044     0     6     3     1     1
11 100051     0    NA    NA    NA    NA
12 100051     0     4     4     2     2
13 100074     0     4     6     4     3
14 100074     0     5     6     3     2
15 100075     0     2     2     1     1

AIM: I need to aggregate by ID (t is always equal to 0) for each variable from P1,P2,P3,P4 like this:

new_data<-aggregate(P1~ID+t,data,mean,na.rm=T)
new_data<-aggregate(P2~ID+t,data,mean,na.rm=T)
new_data<-aggregate(P3~ID+t,data,mean,na.rm=T)
new_data<-aggregate(P4~ID+t,data,mean,na.rm=T)

PROBLEM: Is there a loop I can run or some code from the apply family instead of going through each variable (P1-P4) manually. Thanks a lot!

Answer 1

Haven't tested it, but this should do the loop:

cols<-c("P1","P2","P3","P4")
dat2<-lapply(data[cols],function(x){
  aggregate(x~ID+t, data, mean, na.rm=T)
})

Answer 2

You can aggregate multiple variables at once with cbind(P1, P2, P3, P4) ~ ID + t or equivalently using a dot in place of cbind(P1, P2, P3, P4) . The dot means every remaining variable.

> aggregate(. ~ ID + t, old.data, mean,na.rm=T)
      ID t  P1  P2  P3  P4
1 100003 0 5.5 3.0 2.0 2.5
2 100013 0 5.0 5.0 5.5 2.0
3 100014 0 1.0 1.0 1.0 1.0
4 100015 0 6.0 6.0 1.0 1.0
5 100044 0 6.0 2.5 3.0 1.0
6 100051 0 4.0 4.0 2.0 2.0
7 100074 0 4.5 6.0 3.5 2.5
8 100075 0 2.0 2.0 1.0 1.0
>