Trouble plotting dates with ggplot2

Question

I'm trying to plot against dates in R. I've run into trouble trying to create vertical lines against a plot that I already have. All of the different formats that I try either result in nothing showing up on the plot, or a line at 1970 (the default date). The year-data is in the form yyyy-mm-dd. For example, "1914-07-01".

I've also tried inputting these dates in a data.frame, but got the same problem.

I've been trying to make a reproducible example, but I haven't seen any example datasets to do so with, and got frustrated trying to create one... sorry about that. Here's the relevant code:

ggplot(M,aes(x=date,color=origin,y=value)) + 
  geom_point() +
  geom_line() +
  facet_grid(topic~origin) +
  geom_vline(xintercept=as.numeric(as.Date("1914-07-01")))

Everything plots correctly without the addition of the final line.

Edit: here's the result of dput(head(M)):

structure(list(topic = structure(c(1L, 1L, 1L, 1L, 1L, 1L), .Label = c("1", 
"2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", 
"14", "15", "16", "17", "18", "19", "20", "21", "22", "23", "24", 
"25"), class = "factor"), date = structure(c(-1767196800, -1765987200, 
-1764518400, -1763308800, -1762099200, -1760889600), class = c("POSIXct", 
"POSIXt"), tzone = ""), origin = structure(c(2L, 2L, 2L, 2L, 
2L, 2L), .Label = c("Blast", "The_Egoist"), class = "factor"), 
    value = c(6.69960398194253e-07, 7.48757156068349e-07, 7.04834977806836e-07, 
    7.10226526475778e-07, 6.8295233938925e-07, 6.16466066169137e-07
    )), .Names = c("topic", "date", "origin", "value"), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), row.names = c(NA, -6L), vars = list(
    topic, date), drop = TRUE, indices = list(0L, 1L, 2L, 3L, 
    4L, 5L), group_sizes = c(1L, 1L, 1L, 1L, 1L, 1L), biggest_group_size = 1L, labels = structure(list(
    topic = structure(c(1L, 1L, 1L, 1L, 1L, 1L), .Label = c("1", 
    "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", 
    "13", "14", "15", "16", "17", "18", "19", "20", "21", "22", 
    "23", "24", "25"), class = "factor"), date = structure(c(-1767196800, 
    -1765987200, -1764518400, -1763308800, -1762099200, -1760889600
    ), class = c("POSIXct", "POSIXt"), tzone = "")), class = "data.frame", row.names = c(NA, 
-6L), .Names = c("topic", "date"), vars = list(topic, date)))

for reproducible data, why not give the first few lines of your data: dput(head(M)) — jeremycg
– jeremycg, Commented Aug 6, 2015 at 1:10

jeremycg · Accepted Answer · 2015-08-06 02:00:08Z

3

You were very close, the problem is your data is in POSIXct, and you were trying to convert to Date. To fix it, change to POSIXct:

ggplot(M,aes(x=date,color=origin,y=value)) + 
  geom_point() +
  geom_line() +
  facet_grid(topic~origin) +
  geom_vline(xintercept=as.numeric(as.POSIXct("1914-07-01")))

You can see the difference in the calls:

as.numeric(as.POSIXct("1914-07-01"))
[1] -1751569200

as.numeric(as.Date("1914-07-01"))
[1] -20273

Explaining why the intecept was so close to 1970 (the 0 for both)

answered Aug 6, 2015 at 2:00

jeremycg

25k6 gold badges67 silver badges77 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

mlinegar Over a year ago

This worked! Thank you! If I may, what makes this data POSIXct rather than a date? I was under the impression that in order for it to be POSIXct, there had to be a time attached.

jeremycg Over a year ago

They are both ways of holding dates - POSIX time is seconds since 01-01-1970, date is in days since 01-01-1970. Your original data was in POSIX, so that's what ggplot used. When you gave it an as.numeric(date), it thought you meant seconds before 1970, not days, leading to the vline being near to 1970 instead of where you wanted it.

mlinegar Over a year ago

Ah, that makes sense. But to clarify - I was more wondering what made the date POSIX rather than a date. Was it the format, or is that just the default? Or something else?

jeremycg Over a year ago

impossible for me to tell, str(M$date) gives it as POSIXct - your source for the data must have determined it. I think R normally treats anything with seconds as POSIXct but I could be wrong.

Collectives™ on Stack Overflow

Trouble plotting dates with ggplot2

1 Answer 1

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related