用来自较早观察的data.frame的数据填充data.frame

Question

I have 2 data.frames with observations, mostly the same plots, but one is from this year and one from 2012. I'm using R studio on Windows7. What I want: Create a new column in the new data.frame with the diameter of the trees 5 years ago How I want it: R shall compare the 2 data.frames, and If the location and the tree ID mathes in both it shall copy the diameter of the 2012's data.frame in the new created column of the recent one. My code so far is:

df17$dbh12[df17$LOC=="1"] <- ifelse((df12$ID[df12$LOC=="1"]) %in% (df17$ID[df17$LOC=="1"]), df12$DBH[df12$LOC=="1"], NA)

My Problem is: R is doing it. But, the two data.frames are not identical. In 2012, some of the trees hasn't been considered because they looked sick, but now they are still alive and I measured them. Instead, other trees are dead. I have 10 plots. As example, my data and my code looks like:

df2012=data.frame(LOC=1, ID=c(1,2,4,5,6), DBH=c(7.0, 7.5, 10.25, 14.5, 6.75))
df2017=data.frame(LOC=1, ID=c(2,3,4,5,6), DBH=c(7.8, 28.7, 10.3, 13.7, 7.8))

df2017$dbh12[df2017$LOC=="1"] <- ifelse((df2012$ID[df2012$LOC=="1"]) %in% (df2017$ID[df2017$LOC=="1"]), df2012$DBH[df2012$LOC=="1"], NA)

So at the end I have

> df2017
  LOC ID  DBH dbh12
    1  2  7.8    NA
    1  3 28.7  7.50
    1  4 10.3 10.25
    1  5 13.7 14.50
    1  6  7.8  6.75

My Questions: Why tree 2 has no dbh? Why tree 3 has a dbh? Is R just copying them whatever the ID is? Where is my mistake?

Answer 1

We can do a join on , "LOC" and "ID"

library(data.table)
setDT(df2017)[df2012, dbh12 := i.DBH, on = .(LOC, ID)]
df2017
#   LOC ID  DBH dbh12
#1:   1  2  7.8  7.50
#2:   1  3 28.7    NA
#3:   1  4 10.3 10.25
#4:   1  5 13.7 14.50
#5:   1  6  7.8  6.75

---

In the OP' code, it is only subsetting 'ID' based on the 'LOC' value and is not match ing the 'ID' between the two datasets. The %in% returns a logical vector and if the order is not correct and it will assign the values based on the order of occurrence of values and not on the actual matching of 'ID'

So, here we can use match

i1 <- with(df2017, match(ID[LOC==1], with(df2012, ID[LOC==1])))
df2017$dbh12 <- df2012$DBH[i1]
df2017$dbh12
#[1]  7.50    NA 10.25 14.50  6.75