I made small, reproducible example, but my real data is really huge...and have lots of levels.
roomnumber <- c(1,2,3)
color <- c('red','blue','green')
weight <- c(0.1,0.3,0.5)
data <- expand.grid(roomnumber,color,weight)
names(data) <- c('roomnumber','color','weight')
so, the table is as follows. it contains every combinations.
roomnumber color weight
1 1 red 0.1
2 2 red 0.1
3 3 red 0.1
4 1 blue 0.1
5 2 blue 0.1
6 3 blue 0.1
7 1 green 0.1
8 2 green 0.1
9 3 green 0.1
10 1 red 0.3
11 2 red 0.3
12 3 red 0.3
13 1 blue 0.3
14 2 blue 0.3
15 3 blue 0.3
16 1 green 0.3
17 2 green 0.3
18 3 green 0.3
19 1 red 0.5
20 2 red 0.5
21 3 red 0.5
22 1 blue 0.5
23 2 blue 0.5
24 3 blue 0.5
25 1 green 0.5
26 2 green 0.5
27 3 green 0.5
however, in this data, I want something more. when the "color" variable is "blue", I want to add more combination
specifically, the weight is either 0.1, 0.3, or 0.5 for this combinations but when the "color" is "blue", I want a weight to be 0.1, 0.3, 0.5 AND 0.7
weight_2 <- c(0.1, 0.3, 0.5, 0.7)
how do I edit this?
(I can easily do manually with this minimal example, but my real data is so huge that it is hard to do manually)
There are a couple of ways, both slightly inefficient (though that may be unavoidable).
Add it for all, then remove 0.7 from non-blue rows.
data <- expand.grid(roomnumber = 1:3, color = c("red", "blue", "green"), weight = c(0.1, 0.3, 0.5, 0.7))
data <- subset(data, color == "blue" | weight != 0.7)
data
# roomnumber color weight
# 1 1 red 0.1
# 2 2 red 0.1
# 3 3 red 0.1
# 4 1 blue 0.1
# 5 2 blue 0.1
# 6 3 blue 0.1
# 7 1 green 0.1
# 8 2 green 0.1
# 9 3 green 0.1
# 10 1 red 0.3
# 11 2 red 0.3
# 12 3 red 0.3
# 13 1 blue 0.3
# 14 2 blue 0.3
# 15 3 blue 0.3
# 16 1 green 0.3
# 17 2 green 0.3
# 18 3 green 0.3
# 19 1 red 0.5
# 20 2 red 0.5
# 21 3 red 0.5
# 22 1 blue 0.5
# 23 2 blue 0.5
# 24 3 blue 0.5
# 25 1 green 0.5
# 26 2 green 0.5
# 27 3 green 0.5
# 31 1 blue 0.7
# 32 2 blue 0.7
# 33 3 blue 0.7
rbind the additional blue-only rows.
data <- expand.grid(roomnumber = 1:3, color = c("red", "blue", "green"), weight = c(0.1, 0.3, 0.5))
data <- rbind(data, expand.grid(roomnumber = 1:3, color = "blue", weight = 0.7))
data
# roomnumber color weight
# 1 1 red 0.1
# 2 2 red 0.1
# 3 3 red 0.1
# 4 1 blue 0.1
# 5 2 blue 0.1
# 6 3 blue 0.1
# 7 1 green 0.1
# 8 2 green 0.1
# 9 3 green 0.1
# 10 1 red 0.3
# 11 2 red 0.3
# 12 3 red 0.3
# 13 1 blue 0.3
# 14 2 blue 0.3
# 15 3 blue 0.3
# 16 1 green 0.3
# 17 2 green 0.3
# 18 3 green 0.3
# 19 1 red 0.5
# 20 2 red 0.5
# 21 3 red 0.5
# 22 1 blue 0.5
# 23 2 blue 0.5
# 24 3 blue 0.5
# 25 1 green 0.5
# 26 2 green 0.5
# 27 3 green 0.5
# 28 1 blue 0.7
# 29 2 blue 0.7
# 30 3 blue 0.7