R: expand sequence of binary values from two columns - r

Question

I have a very complex dataset. But for simplification I am going to use this data. Let's assume we have this data, one column X showing numbers in different ranges and the other column is binary 0/1. I want to expand a sequence of binary 0/1 by 1. This will create new longer columns (newX, newbinary) form the original X and binary columns.
X binary
1 1 1
2 3 0
3 5 0
4 6 1
5 8 1
6 10 0
I want to expand X column here and put the binary value next to it. Something like,
newX newbinary
1 1 1
2 2 1
3 3 0
4 4 0
5 5 0
6 6 1
7 7 1
8 8 1
9 9 1
10 10 0
My attempts is this, First I created the new X column (expanded the X).
newX <- seq(X[1], X[length])
Then, I used for loop to iterate through newX and then compare the value with X to check if it is equal to or less than the value. If newX[i] is not equal to X[i], then put the previous binary value, else put the binary[i].
for (i in 1:newX[length])
{
newbinary= ifelse((newX != X)&(between(newX[i], X[i],X[j+1])), lag(binary), binary)
}
But this is not working and give me this,
newX newbinary
1 1 1
2 2 NA
3 3 NA
4 4 NA
5 5 NA
6 6 NA
I don't really know how the newX column (longer one) will iterate through X (shorter one) and put the value correspondingly.
How can I achieve this in R?

Answer 1

Try this tidyverse approach creating a dataframe for the sequence, then use left_join() and finally fill() to complete the binary variable:
library(dplyr)
library(tidyr)
#Code
newdf <- data.frame(X=seq(min(df$X),max(df$X),by=1)) %>%
left_join(df) %>%
fill(binary)
Output:
X binary
1 1 1
2 2 1
3 3 0
4 4 0
5 5 0
6 6 1
7 7 1
8 8 1
9 9 1
10 10 0
Some data used:
#Data
df <- structure(list(X = c(1L, 3L, 5L, 6L, 8L, 10L), binary = c(1L,
0L, 0L, 1L, 1L, 0L)), class = "data.frame", row.names = c("1",
"2", "3", "4", "5", "6"))

Answer 2

Does this work:
library(dplyr)
library(tidyr)
library(purrr)
df %>% mutate(y = lead(X, default = max(X)+1), newX = map2(X,y-1,`:`)) %>%
unnest(newX) %>% mutate(newbinary = binary) %>% select(newX, newbinary)
# A tibble: 10 x 2
newX newbinary
<int> <dbl>
1 1 1
2 2 1
3 3 0
4 4 0
5 5 0
6 6 1
7 7 1
8 8 1
9 9 1
10 10 0
Data used:
df
# A tibble: 6 x 2
X binary
<dbl> <dbl>
1 1 1
2 3 0
3 5 0
4 6 1
5 8 1
6 10 0

Answer 3

Assuming the data frame shown reproducibly in the Note at the end the following one-linear uses read.zoo to convert it to a zoo object and converts that to a ts object expanding it at the same time. Then it uses na.locf to fill in the NAs and fortify.zoo to convert it to a data.frame. Set the names on the data frame to the original names.
library(zoo)
setNames(fortify.zoo(na.locf(as.ts(read.zoo(DF)))), names(DF))
giving:
X binary
1 1 1
2 2 1
3 3 0
4 4 0
5 5 0
6 6 1
7 7 1
8 8 1
9 9 1
10 10 0
The above one-linear could alternately be written as a pipeline:
library(zoo)
library(magrittr)
DF %>%
read.zoo %>%
as.ts %>%
na.locf %>%
fortify.zoo %>%
setNames(names(DF))
Note
Lines <- "
X binary
1 1 1
2 3 0
3 5 0
4 6 1
5 8 1
6 10 0"
DF <- read.table(text = Lines)