CODEWITHSUNDEEP
rdateif-statementlubridate
Hendrik
Hendrik

Reputation: 135

Date manipulation with lubridate and if-statement

I am struggling with some data manipulation. One of the columns in my datasheet contains the date of birth, but for one location the values are off by 100 years.

I made an example small data frame to explain my problem: the dates for Paris / Berlin are correct, I want to change the date only for those rows with London as location (for this example from 2028-3-25 to 1928-3-25).

library(lubridate)
date <- as.Date(c('1950-11-1','2028-3-25','1940-3-14'))
location <- c("Paris", "London", "Berlin")
df <- data.frame(date, location)
df$date_new <- ifelse(df$location %in% c("London"), df$date - years(100), df$date)

As you can see, I installed the lubridate package and tried to use an if else statement, but that just gives me some negative numbers in the new column.

The solution is probably very simple, but I cannot figure it out and it's driving me insane.

Thank you!

Upvotes: 4

Views: 643

Answers (2)

Se&#241;or O
Se&#241;or O

Reputation: 17412

ifelse is taking the class attributes from the test:

The mode of the result may depend on the value of test (see the examples), and the class attribute (see oldClass) of the result is taken from test and may be inappropriate for the values selected from yes and no.

Sometimes it is better to use a construction such as

(tmp <- yes; tmp[!test] <- no[!test]; tmp) , possibly extended to handle missing values in test.

So it looks like it's best not to use ifelse. Here's one solution:

> df$date_new = df$date
> df[location == "London",]$date_new = df[location == "London",]$date_new - years(100)
> df
        date location   date_new
1 1950-11-01    Paris 1950-11-01
2 2028-03-25   London 1928-03-25
3 1940-03-14   Berlin 1940-03-14

However, if you want to use ifelse, you can coerce the object into a Date if you specify the standard origin (an object in R)

> library(lubridate)
> date <- as.Date(c('1950-11-1','2028-3-25','1940-3-14'))
> location <- c("Paris", "London", "Berlin")
> df <- data.frame(date, location)
> df$date_new <- as.Date(ifelse(df$location == "London", as.Date(df$date - years(100)), df$date), origin = origin)
> df
        date location   date_new
1 1950-11-01    Paris 1950-11-01
2 2028-03-25   London 1928-03-25
3 1940-03-14   Berlin 1940-03-14

Upvotes: 3

dimitris_ps
dimitris_ps

Reputation: 5951

Try this as an alternative

df$date_new <- df$date
df$date_new[df$location=="London"] <- df$date_new[df$location=="London"] - years(100)

or instead of df$date_new <- ifelse(df$location %in% c("London"), df$date - years(100), df$date)

try 

df$date_new <- ifelse(df$location %in% c("London"), as.character(df$date - years(100)), as.character(df$date))

Upvotes: 4

Related Questions