Error message plotting timeseries with ggplot2

Question

I am very new to ggplot2 and wondered whether anyone could help me with my simple problem. My sample dataframe is given as follows. I want to plot the time (hours and minutes) against the number of Counts.

Date <- c("07/12/2012 05:00:00", "07/12/2012 06:00:00", "07/12/2012 07:00:00",
      "07/12/2012 08:00:00")
Date <- strptime(Date, "%d/%m/%Y %H:%M")
Counts <- c("0","3","10","6")
Counts <- as.numeric(Counts)
df1 <- data.frame(Date,Counts,stringsAsFactors = FALSE)

#Adds time to dataframe.
df1 <- within(df1,{
  posb <- as.POSIXlt(Date,format="%d/%m/%Y %H:%M")
hours <- posb$hour
mins <- posb$min
dates <- format(posb, "%x")
time <- format(posb, "%H:%M")
posb <- NULL  # cleanup
})
#Draw graph (time versus Counts)
library(ggplot2)
g = ggplot(df1, aes(x=time, y=Counts))
g + geom_line()

I always get the error message 'geom_path: Each group consist of only one observation. Do you need to adjust the group aesthetic?'.

Can anyone help me rectify my code to allow my graph to plot correctly?

Edit I am still trying to get the formatting of my time variable to work, and plot a graph. However at the moment, it's still not recognising the date format. I would like to be able to: 1. Plot a set period of data (say from 07:47:50 to 07:49:10). 2. Ask R to plot the x axis every whole minute. ... neither of these I can get to work at the moment. A subset of my real data is shown below. Any advice would be gratefully received.

day3 <- structure(list(Date = c("11/12/2012", "11/12/2012", "11/12/2012", 
                            "11/12/2012", "11/12/2012", "11/12/2012", "11/12/2012", "11/12/2012", 
                            "11/12/2012", "11/12/2012", "11/12/2012", "11/12/2012", "11/12/2012", 
                            "11/12/2012", "11/12/2012", "11/12/2012", "11/12/2012", "11/12/2012", 
                            "11/12/2012", "11/12/2012"), Time = c("07:46:10", "07:46:20", 
                                                                  "07:46:30", "07:46:40", "07:46:50", "07:47:00", "07:47:10", "07:47:20", 
                                                                  "07:47:30", "07:47:40", "07:47:50", "07:48:00", "07:48:10", "07:48:20", 
                                                                  "07:48:30", "07:48:40", "07:48:50", "07:49:00", "07:49:10", "07:49:20"
                            ), Axis1 = c(59L, 651L, 59L, 0L, 22L, 50L, 0L, 0L, 114L, 899L, 
                                         129L, 33L, 21L, 9L, 224L, 135L, 266L, 16L, 59L, 126L), Steps = c(1L, 
                                                                                                          2L, 1L, 0L, 2L, 1L, 0L, 0L, 5L, 15L, 6L, 2L, 2L, 0L, 8L, 5L, 
                                                                                                          16L, 1L, 3L, 8L)), .Names = c("Date", "Time", "Axis1", "Steps"
                                                                                                          ), row.names = 52838:52857, class = "data.frame")
#Creates a new dataframe with a time column.
day3 <- within(day3,{
  posb <- as.POSIXlt(Time,format="%H:%M:%S")
  posb <- NULL  # cleanup
})

library(ggplot2)
g = ggplot(day3, aes(x=strptime(Time, "%H:%M:%S"), y=Axis1)) + geom_line(aes(group = 1)) +
  theme_bw() +
  xlab("Time") + 
  ylab("Activity (Counts per 10 seconds)") + 
  scale_x_datetime(limits=c(as.POSIXct("07:47:50"),as.POSIXct("07:49:10")))


g

juba juba · Accepted Answer · 2013-02-11T14:50:15

The problem is due to the fact that your time variable is a character vector :

R> class(df1$time)
[1] "character"

You must convet it to an object of class POSIXlt, for example like this :

ggplot(df1, aes(x=strptime(time, "%H:%M"), y=Counts)) + geom_line()

Or, much simpler, you can directly use your Date variable, without transformation :

ggplot(df1, aes(x=Date, y=Counts)) + geom_line()

enter image description here

Better yet, you will see that ggplot2 automagically labels your x axis depending on your timespan along the axis.

EDIT : if you want to define the x-axis limits, you can do something like :

ggplot(df1, aes(x=Date, y=Counts)) + geom_line() + scale_x_datetime(limits=c(as.POSIXct("2012/12/07 04:00:00"),as.POSIXct("2012/12/07 10:00:00")))

Error message plotting timeseries with ggplot2

1 Answers