CODEWITHSUNDEEP
rdplyrtidyversezoo
ah bon
ah bon

Reputation: 10051

Interpolate certain columns' NAs based on the condition of another column using R

Given a sample data as follows:

df <- structure(list(date = structure(c(18912, 18913, 18914, 18915, 
18916, 18917, 18918, 18919, 18920, 18921, 18922, 18923), class = "Date"), 
    value1 = c(1.115668, NA, 
    1.093685, NA, 1.072135, 1.06152, 1.05101, NA, NA, 1.0201, 
    1.01, 1), value2 = c(1.015, 1.030225, NA, NA, 1.077284, 1.093443, 
    1.109845, 1.126493, NA, NA, NA, 1.195618
    )), row.names = c(NA, -12L), class = "data.frame")

Let's say I need to interplote NAs in value1 and value2 but only if date is in the date range of date>='2021-10-16' & date<='2021-10-20'.

I used this code but it was apply to the whole column value1.

library(zoo)
df$value1 <- na.approx(df$value1)

How could I achieve that based on code above? Thanks.

Upvotes: 0

Views: 56

Answers (1)

Sotos
Sotos

Reputation: 51592

You can filter your period and loop over the columns applying the function, i.e.

sapply(df[-1], function(i)zoo::na.approx(i[df$date>='2021-10-16' & df$date<='2021-10-20']))

Upvotes: 1

Related Questions