Making new variables for every group of observation in R

问题

I have 11 variables in my dataframe. The first is unique identifier of observation (a plane). The second one is a number from 1 to 21 representing flight of a given plane. The rest of the variables are time, velocity, distance, etc.

What I want to do is make new variables for every group (number) of flight e.g. time_1, time_2,..., velocity_1, velocity_2, etc. and consequently, reduce the number of observations (the repeating ones).

I don't really have idea how to start. I was thinking about a mutate function like:

mutate(df, time_1 = ifelse(n_flight == 1, time, NA))

But that would be a lot of typing and a new problem may appear, perhaps.

回答1:

Basically, you want to convert long to wide data for each variable. You can lapply over these with tidyr::spread in that case. Suppose the data looks like the following:

library(dplyr)
library(tidyr)
df <- data.frame(
  ID = c(rep("A", 3), rep("B", 3)), 
  n_flight = rep(seq(3), 2),
  time = seq(19, 24), 
  velocity = rev(seq(65, 60))
)

Then the following will generate your outcome of interest, as long as you get rid of the extra ID variables.

lapply(
  setdiff(names(df), c("ID", "n_flight")), function(x) {
    df %>% 
      select(ID, n_flight, !!x) %>%
      tidyr::spread(., key = "n_flight", value = x) %>%
      setNames(paste(x, names(.), sep = "_"))
  }
) %>%
  bind_cols()

Let me know if this wasn't what you were going for.

来源：https://stackoverflow.com/questions/55463593/making-new-variables-for-every-group-of-observation-in-r