R: plot multiple curves vs one var but for 4 factors

Question

I have a DF that looks like:

   id     app  vac dac
1:  1 1000802  579 455
2:  1 1000803 1284 918
3:  1 1000807   68  66
4:  1 1000809 1470 903
5:  2 1000802  407 188
6:  2 1000803  365 364
7:  2 1000807  938 116
8:  2 1000809  699 570

I need to plot vac and dac for each app on same canvas as a function of id. I know how to do it for only one app by using melt and bulk-plot with ggplot. But I'm stuck how to do it for arbitrary number of factors/levels.

In this example there will be total 8 curves for 4 app. Any thoughts? Here's the data frame for tests. Thank you!!

df = structure(list(id = c(1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L), app = c(1000802, 
                                                               1000803, 1000807, 1000809, 1000802, 1000803, 1000807, 1000809
), vac = c(579, 1284, 68, 1470, 407, 365, 938, 699), dac = c(455, 
                                                             918, 66, 903, 188, 364, 116, 570)), .Names = c("id", "app", "vac", 
                                                                                                            "dac"), class = c("data.table", "data.frame"), row.names = c(NA, 
                                                                                                                                                                         -8L))

Edit: some clarification on axes, x axis = id, y axis = values of vac and dac for each of 4 app factors

score 2 · Accepted Answer · answered Feb 07 '17 at 17:52

It is a bit unclear what you are looking for, but if you are looking for a line connecting the values of vac and dac, here is a solution using dplyr and tidyr.

First, gather the vac and dac columns (this is similar to reshape2::melt but with a syntax I find easier to follow). Then, set the variable (which has "vac" and "dac") as your x-locations, the value (from the old vac and dac columns) as your y and then map app and id to aesthetics (here, color and linetype). Set the group to ensure that it connects the right pairs of points, and add geom_line:

df %>%
  gather(variable, value, vac, dac) %>%
  ggplot(aes(x = variable
             , y = value
             , color = factor(app)
             , linetype = factor(id)
             , group = paste(app, id))) +
  geom_line()

gives

Given the question edit, you can change axes like so:

df %>%
  gather(variable, value, vac, dac) %>%
  ggplot(aes(x = id
             , y = value
             , color = factor(app)
             , linetype = variable
             , group = paste(app, variable))) +
  geom_line()

gives

2nd solution is exactly what I was looking for. Perfect! Thank you very much! — Alexey Ferapontov, Feb 07 '17 at 17:57

score 1 · Answer 2 · answered Feb 07 '17 at 17:50

1

I not sure, I understood your question but I would do something like

ggplot(df,aes(vac,app,group=app)) + geom_point(aes(color=factor(app)))

answered Feb 07 '17 at 17:50

DJJ

2,481
2
28
53

Not exactly this, but thanks. `x` axis = `id`, `y` axis = values of `vac` and `dac` for each of 4 `app` factors – Alexey Ferapontov Feb 07 '17 at 17:52

R: plot multiple curves vs one var but for 4 factors

2 Answers2