Replace NA with Last Observed Value in R (Example)

In this R tutorial you’ll learn how to replace NAs with the previous non-NA value.

Table of contents:

• Creation of Example Data
• Example: Replace NA with latest non-NA Value in R
• Video, Further Resources & Summary

Let’s dive into it.

Creation of Example Data

For the example of this tutorial, we will use a very simple numeric vector:

x <- c(1, 2, NA, NA, NA, 3, NA, 4)    # Create example data
x                                     # Print example data
# 1  2 NA NA NA  3 NA  4

As you can see based on the previous R code, our example vector contains values ranging from 1 to 4 with NA values in between.

Example: Replace NA with latest non-NA Value in R

Let’s assume that we want to replace these NA values with the latest observed value in our data vector. We can do that based on the zoo add-on package. Let’s install and load the package to RStudio:

install.packages("zoo")               # Install zoo package
library("zoo")                        # Load zoo package

The zoo R package contains the na.locf function, which is a generic function for replacing each NA with the most recent non-NA value prior to it. Let’s do this in practice:

na.locf(x)                            # Apply na.locf function
# 1 2 2 2 2 3 3 4

Based on the previous output of the RStudio console you can see that all NA values were replaced with the previously observed number.

Video, Further Resources & Summary

I have recently published a video on my YouTube channel, which explains the R programming codes of this article. You can find the video below:

Besides that, you might want to have a look at the related articles on my website. Please find a selection of interesting articles here:

• Replace NA with 0
• coalesce Function of dplyr Package
• R is.na Function
• R na_if Function of dplyr Package
• What are NA Values?
• Remove NA Values from Vector in R
• R Functions List (+ Examples)
• The R Programming Language

In summary: In this R tutorial you learned how to fill missing values using the previous observation. In case you have further questions, let me know in the comments section below.