Is it possible to extend a database based on the last value?
For example:
x <- c(1, 2, 3, 4, 5)
y <- c(0.4, 0.6, 0.2, 0.2, 0.1)
df <- cbind(x, y)
I would like to be able to extend x to 10 and just use the last y value (0.1) to be populated all the way (i.e. in all the missing fields x = 6, 7, 8, 9, 10). Is it possible?
Here is a base R idea,
rbind(df, setNames(data.frame(x = seq(6, 10), y = df$y[nrow(df)]), names(df)))
# x y
#1 1 0.4
#2 2 0.6
#3 3 0.2
#4 4 0.2
#5 5 0.1
#6 6 0.1
#7 7 0.1
#8 8 0.1
#9 9 0.1
#10 10 0.1
You can use complete and fill from tidyr :
library(tidyr)
df %>% complete(x = 1:10) %>% fill(y)
# A tibble: 10 x 2
# x y
# <dbl> <dbl>
# 1 1 0.4
# 2 2 0.6
# 3 3 0.2
# 4 4 0.2
# 5 5 0.1
# 6 6 0.1
# 7 7 0.1
# 8 8 0.1
# 9 9 0.1
#10 10 0.1
Make data as dataframe :
df <- data.frame(x, y)
Not as elegant as the other answer, but you can use zoo::na.locf (last observation carried forward) for similar problems:
library(zoo)
x <- 1:5
y <- c(0.4, 0.6, 0.2, 0.2, 0.1)
df <- data.frame(
x = 1:10,
y = NA_real_
)
df[1:length(x), ]$y <- y
df$y <- na.locf(df$y)
Continuing with @RonakShah's answer, you can also use the fill argument in complete() to replace missing values.
library(tidyr)
df %>% complete(x = 1:10, fill = list(y = y[nrow(.)]))
# # A tibble: 10 x 2
# x y
# <dbl> <dbl>
# 1 1 0.4
# 2 2 0.6
# 3 3 0.2
# 4 4 0.2
# 5 5 0.1
# 6 6 0.1
# 7 7 0.1
# 8 8 0.1
# 9 9 0.1
# 10 10 0.1
or
df %>% complete(x = 1:10, fill = list(y = dplyr::last(y)))
A base R option using approx
data.frame(with(df, approx(x, y, xout = 1:10, rule = 2)))
such that
> data.frame(with(df, approx(x, y, xout = 1:10, rule = 2)))
x y
1 1 0.4
2 2 0.6
3 3 0.2
4 4 0.2
5 5 0.1
6 6 0.1
7 7 0.1
8 8 0.1
9 9 0.1
10 10 0.1