ggplot2 multiple lines on a single graph

Question

I have a data set that runs for 7 days producing cumulative counts for each of those days broken up into 15 min periods

The time period starts from 12 o'clock in the morning and runs until 12 o clock the following day. The order is important.

Below is a made up sample (apologies for the messy nature)

library(tidyverse)

add_break <- function(check) {
      Zero_break <- paste(check, '00', sep=":") %>% as_tibble()
      fifteen_break <- paste(check, '15', sep=":") %>% as_tibble()
      thirty_break <- paste(check, '30', sep=":") %>% as_tibble()
      fortyfive_break <- paste(check, '45', sep=":") %>% as_tibble()
      bind_rows(Zero_break, fifteen_break, thirty_break, fortyfive_break)
}

# Create the Levels for the Time for every 15 mins
pm <- paste(seq(12,23), sep='') %>% as_tibble()
am <- paste(0, seq(00,09), sep='') %>% as_tibble()
am_2 <- paste(seq(10,11), sep='') %>% as_tibble()
clock <- pm %>% bind_rows(am, am_2)
intervals <- map_df(clock$value, add_break)

# Create Random data for cumsum
mydf <- intervals %>% 
  mutate(MON = cumsum(sample(1:100,size = 96,replace = TRUE)),
         TUE = cumsum(sample(1:100,size = 96,replace = TRUE)),
         WED = cumsum(sample(1:100,size = 96,replace = TRUE)),
         THUR = cumsum(sample(1:100,size = 96,replace = TRUE)),
         FRI = cumsum(sample(1:100,size = 96,replace = TRUE)),
         SAT = cumsum(sample(1:100,size = 96,replace = TRUE)),
         SUN = cumsum(sample(1:100,size = 96,replace = TRUE)))

mydf$AVG <- round(rowMeans(mydf[,2:8]),2)

ggplot(mydf, aes(x=reorder(value, MON), y = MON)) +
geom_line() +
geom_line(aes(x=reorder(value, TUE), y = TUE), colour = 'red')

I then try and create a line graph where i want to create very light colours for the days of the week and then a pretty strong colour for the average

Unfortunately when i run the code i get the following error

geom_path: Each group consists of only one observation. Do you need to adjust the group aesthetic? geom_path: Each group consists of only one observation. Do you need to adjust the group aesthetic?

Can Anyone help?

bVa · Accepted Answer · 2018-01-31 09:15:49Z

3

Another suggestion with a Date format for x axis, and different alpha levels for each variable value:

# Add a new x variable (supposed to be an hour)
mydf$hour <- as.POSIXct(ifelse(as.double(substr(mydf$value, 1, 2)) >= 12, 
                           paste("2018-01-30 ", mydf$value, ":00", sep = ""), 
                           paste("2018-01-31 ", mydf$value, ":00", sep = "")))


# change to long data format
mydf_m <- gather(mydf, "day", "val", c(2:9))

# reorder variable values
mydf_m$day <- factor(mydf_m$day, levels=c("MON", "TUE", "WED", "THUR", "FRI", "SAT", "SUN", "AVG"))

library(scales) # for date_format in scale_x_datetime
ggplot(mydf_m, aes(x = hour, y = val, color = day, alpha = day, size = day)) +
  theme_bw() + # 
  geom_line() +
  scale_x_datetime(labels = date_format("%H:%M", tz = "CET")) + # x axis
  scale_alpha_manual(values = c("MON" = 0.3, "TUE" = 0.3, "WED" = 0.3, "THUR" = 0.3, "FRI" = 0.3, "SAT" = 0.3, "SUN" = 0.3, "AVG" = 0.8), guide = "none") + 
  scale_size_manual(values = c("MON" = 1, "TUE" = 1, "WED" = 1, "THUR" = 1, "FRI" = 1, "SAT" = 1, "SUN" = 1, "AVG" = 3))

edited Jan 31, 2018 at 9:15

answered Jan 30, 2018 at 16:35

bVa

3,9681 gold badge16 silver badges22 bronze badges

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

5 Comments

John Smith Over a year ago

Hi @bVa, thank you for your answer. Unfortunately the order of the hours is important. It needs to be from 12 mid day to 12 mid day. I tried converting everything to a time using lubridate hm function but ggplot (quite rightly) sorts it based on ascending time

bVa Over a year ago

@JohnSmith I edited my answer with a little trick to have a correct x scale.

John Smith Over a year ago

Hi @bVa, when running the code to create the long dataframe i get Error: Column "value" must have a unique name

bVa Over a year ago

@JohnSmith Sorry, I forgot to change a variable name : change value to val in gather, I edited it.

John Smith Over a year ago

Perfect, exactly what i was looking for. Im going to be digesting this for the afternoon. Thanks again, you really have gotten me out of a hole

Antonios · Accepted Answer · 2018-01-31 09:32:00Z

2

with some help from this answer https://stackoverflow.com/questions/35586520/when-creating-a-multiple-line-plot-in-ggplot2-how-do-you-make-one-line-thicker

I got the following, which rounds the x-axis to hours:

mydf0=melt(ungroup(mydf))
names(mydf0)[1]="Time"
mydf0$Time=as.factor(as.character(hour(ymd_hm(paste("2018-01-01",mydf0$Time,sep=" ")))))
mydf0$Time <- factor(mydf0$Time,levels=c(as.character(seq(12,23,1)),as.character(seq(0,11,1))))
levels(mydf0$Time)
mydf0$size <- rep(.5, nrow(mydf0))
mydf0$size[mydf0$variable=="AVG"] <- 2
ggplot(mydf0, aes(x=Time, y = value,col=variable,group=variable,size=size))+geom_line() +
  scale_size(range = c(0.5, 2), guide="none")

hope it is helpful

if we do not round the hour the x-axis is not really visible, while lines are smoother

    mydf0=melt(ungroup(mydf))
names(mydf0)[1]="Time"
mydf0$Time <- ordered(mydf0$Time,levels=mydf0$Time[1:96])
mydf0$size <- rep(.5, nrow(mydf0))
mydf0$size[mydf0$variable=="AVG"] <- 2
ggplot(mydf0, aes(x=Time, y = value,col=variable,group=variable,size=size))+geom_line() +
  scale_size(range = c(0.5, 2), guide="none")+ scale_x_discrete(breaks=as.character(seq(0,23,1)))

edited Jan 31, 2018 at 9:32

answered Jan 30, 2018 at 15:45

Antonios

1,9371 gold badge12 silver badges18 bronze badges

2 Comments

John Smith Over a year ago

Hi @Antonis, Thank you for answering so quickly. The problem with this method is that the time orders itself from mid night to mid night, i need it to run from mid day to mid day, so basically the order in the mydf$value portion of the dataframe mydf

Antonios Over a year ago

Hi @John Smith. sorry i did not understand that part. I changed my code and i think now it works as you want. If you find a way to display time values in a visible way in graph2 please let me know :)

Collectives™ on Stack Overflow

ggplot2 multiple lines on a single graph

2 Answers 2

5 Comments

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

5 Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related