Replace month with year in x legend

Question

How is it possible to receive a time line plot with multiple time lines in the same plot like this:

df3 <- structure(list(id = c(6L, 6L, 2L, 3L, 6L, 7L, 3L, 5L, 7L, 
                            4L, 6L, 7L, 7L, 3L, 6L, 6L, 6L, 7L, 6L, 6L, 3L, 6L, 6L, 3L, 6L, 
                            7L, 6L, 6L, 7L, 6L, 7L, 6L, 3L, 5L, 6L), month = c("2011-01", 
                                                                                       "2011-02", "2011-04", "2011-04", "2011-04", "2011-04", "2011-05", 
                                                                                       "2011-05", "2011-05", "2011-06", "2011-06", "2011-06", "2011-07", 
                                                                                       "2011-08", "2011-08", "2011-09", "2011-10", "2011-11", "2012-03", 
                                                                                       "2012-04", "2012-05", "2012-05", "2012-08", "2012-09", "2012-09", 
                                                                                       "2012-09", "2012-10", "2012-11", "2012-11", "2012-12", "2012-12", 
                                                                                       "2013-01", "2013-02", "2013-02", "2013-02"), frq = c(1, 
                                                                                                                                                  1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 1, 1, 
                                                                                                                                                  1, 1, 1, 2, 1, 2, 1, 1, 1, 1, 1, 1, 2)), row.names = c(NA, 35L
                                                                                                                                                  ), class = "data.frame")

Additionally how is it possible to add in the x legend the year evenif the plot frequency is per month? Also, in the group variable how is is possible two show exactly which is the variable and the color of the respective line?

What I tried is this

library(ggplot2)
ggplot(df3,aes(x=month,y=frq,colour=id,group=id)) + geom_line()

Does https://stackoverflow.com/q/10576095/3358272 resolve your question? — r2evans, Jul 06 '20 at 16:12
... and you probably want to set to a Date format: `df3$date <- as.Date(paste(df3$month, "01", sep="-"))` and use that as the x-variable. And to show discrete colours your `id` variable should be a `factor` — user20650, Jul 06 '20 at 16:15

score 1 · Accepted Answer · answered Jul 06 '20 at 16:50

1

Do you mean something like this:

library(ggplot2)
df3$id <- as.factor(df3$id)
ggplot(df3,aes(x=month,y=frq,colour=id,group=id)) + geom_line()

answered Jul 06 '20 at 16:50

Duck

39,058
13
42
84

score 0 · Answer 2 · answered Jul 07 '20 at 12:44

Function to read first (n) characters:

firstfew <- function(x, n){
x=as.character(x)
substr(x, 1, n)}

years=firstfew(df3$month,4)
df3$years=years

ggplot(df3,aes(x=month,y=frq,colour=id,group=id)) + geom_line()+ scale_x_discrete(labels= years)

Facets view:

ggplot(df3,aes(x=month,y=frq,colour=id,group=id)) + geom_line()+theme(legend.position="top",axis.ticks.x=element_blank(),axis.title.x=element_blank(),axis.text.x=element_blank(),panel.spacing.x=unit(0, "lines"),panel.grid.major = element_blank())+facet_wrap(~years,strip.position = "bottom",scales = "free_x")

Replace month with year in x legend

2 Answers2