Using Times in a Factor to Create a Custom Column - r

Question

I have a dataset titled nypd, which has a column titled OCCUR_TIME. This column contains various times (ex: 3:57:00, 10:31:00, 22:15:00, etc.).
I would like to create a custom TIME_OF_DAY column using R; I wrote this code below:
nypd$TIME_OF_DAY <- 'Night'
nypd[nypd$OCCUR_TIME >= 6:00:00 & nypd$OCCUR_TIME < 12:00:00,] <- 'Morning'
nypd[nypd$OCCUR_TIME >= 12:00:00 & nypd$OCCUR_TIME < 16:00:00,] <- 'Afternoon'
nypd[nypd$OCCUR_TIME >= 16:00:00 & nypd$OCCUR_TIME < 20:00:00,] <- 'Evening'
The error I am getting is Error in `[<-.data.frame`(`*tmp*`, nypd$OCCUR_TIME >= "6:00:00" & nypd$OCCUR_TIME < : missing values are not allowed in subscripted assignments of data frames.
I'm new to R so I am not too familiar with the error codes, but I'm thinking the error is due to my values in the OCCUR_TIME column not being read as a "time" type of value, so I can't use any operators.
Could someone please help me figure out where I'm going wrong? Thank you!

Answer 1

First, as the error is saying, you have missing values in your data. Since we don't have your data to work with, let's make up some data to use:
> data(iris)
> iris$Petal.Length[3:5] <- NA
> head(iris)
Sepal.Length Sepal.Width Petal.Length Petal.Width Species
1 5.1 3.5 1.4 0.2 setosa
2 4.9 3.0 1.4 0.2 setosa
3 4.7 3.2 NA 0.2 setosa
4 4.6 3.1 NA 0.2 setosa
5 5.0 3.6 NA 0.2 setosa
6 5.4 3.9 1.7 0.4 setosa
Now, it has a problem with subsetting on Petal.Length because it isn't sure what to do when there are missing values.
> iris[iris$Petal.Length > 1.2 & iris$Petal.Length < 1.5, ] <- 50
Error in `[<-.data.frame`(`*tmp*`, iris$Petal.Length > 1.2 & iris$Petal.Length < :
missing values are not allowed in subscripted assignments of data frames
Also note that when you do this:
nypd[nypd$OCCUR_TIME >= 6:00:00 & nypd$OCCUR_TIME < 12:00:00,] <- 'Morning'
You aren't telling it what variable you want to assign 'Morning' to!
You can add a test for is.na to your boolean, and include the variable name you want to affect:
> iris[!is.na(iris$Petal.Length) & iris$Petal.Length > 1.2 & iris$Petal.Length < 1.5, 'Petal.Length'] <- 50
> head(iris)
Sepal.Length Sepal.Width Petal.Length Petal.Width Species
1 5.1 3.5 50.0 0.2 setosa
2 4.9 3.0 50.0 0.2 setosa
3 4.7 3.2 NA 0.2 setosa
4 4.6 3.1 NA 0.2 setosa
5 5.0 3.6 NA 0.2 setosa
6 5.4 3.9 1.7 0.4 setosa
The advice about learning how to deal with dates and times in R is true, the way you are expressing them here is not right. If they are being read in as a factor, then perhaps however you are reading your data you need to add a stringsAsFactors = FALSE?

Answer 2

We could convert the 'OCCUR_TIME' to Time class with as.ITime from data.table, then do the the comparison
library(dplyr)
library(data.table)
nypd %>%
mutate(OCCUR_TIME = as.ITime(OCCUR_TIME),
TIME_OF_DAY = case_when(between(OCCUR_TIME, as.ITime("06:00:00"),
as.ITime("12:00:00")) ~ "Morning",
between(OCCUR_TIME, as.ITime("12:00:00"),
as.ITime("16:00:00")) ~ "Afternoon",
between(OCCUR_TIME, as.ITime("16:00:00"),
as.ITime("20:00:00")) ~ "Evening", TRUE ~ "Night"))
# OCCUR_TIME TIME_OF_DAY
#1 05:22:34 Night
#2 07:22:29 Morning
#3 12:20:05 Afternoon
#4 15:46:23 Afternoon
#5 19:32:42 Evening
data
nypd <- data.frame(OCCUR_TIME = c("05:22:34", "07:22:29", "12:20:05",
"15:46:23", "19:32:42"), stringsAsFactors = FALSE)