Plotting data against time/date on the x-axis in ggplot2

General
,
1

Im trying to plot a value column against a time column. So far my code looks like this:

ggplot(echo120_edit, aes(y= PRC_ABC, x =Time_M)) + geom_smooth(method = loess, se = TRUE, fullrange= TRUE) +
scale_x_continuous(breaks = seq(0, 23, 1), labels = paste(seq(0, 23, 1), "00", sep = ":")) +
geom_point() + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_rect(fill = "white", colour = NA))

The time column has the time the variable was collected (its area backscattering coefficient data btw). I can plot on against the other, but the x-axis looks like a mess and I cant get the geom_smooth lines to appear when I do that. Essentially, I'm stuck trying to make the x-axis (time) have a repeating 24 hour clock with date marks with the loess line and varibale data plotted against it.

2

What kind of format is the time data in? If it's a time or a datetime, it should automatically use a reasonable time labeling convention.

Are you trying to plot multiple days worth of data (and smoothed data) on the same graph? Or is it one 24-hour section?

3

When this happens to me, I can usually debug it with one or both of two fixes:

  1. Converting the column with the date/time information to the right class (e.g., from string to date).

  2. Using scale_x_date() or scale_x_datetime() in my call to ggplot().

The nice thing with the latter is that it also affords you finer control over the appearance of that axis, e.g. different breaks or nicer labels.

4

the date and time were in 2 columns but I combined them. It was 24 hour periods over ~2 days.

I ultimately managed to convert them to POSIXct and used a cbind function to put it into the data. So far it seems to have worked but needs tweeking.

5

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.