面板数据和行 plot 用于 R 中的几个变量

Question

I have panel data where "time" variable means year and "unit_id" means country. I'd like to make line plot of all numeric variables (values of these variables accross year). One line for all countries of one variable across time.

I tried:

exampl%>%
  gather() %>%                             # Convert to key-value pairs
  ggplot(aes(time)) +                     # Plot the values
  facet_wrap(~ key, scales = "free") +   # In separate panels
  geom_line()

But it doesnt work, I have an error: Aesthetics must be valid data columns. Problematic aesthetic(s): x = time. Did you mistype the name of a data column or forget to add after_stat()

Thats my data:

dput(exampl)
structure(list(unit_id = c("AGO", "AGO", "AGO", "AGO", "AGO", 
"AGO", "BEN", "BEN", "BEN", "BEN", "BEN", "BEN", "BGD", "BGD", 
"BGD", "BGD", "BGD", "BGD", "CIV", "CIV", "CIV", "CIV", "CIV", 
"CIV"), time = c(2005L, 2006L, 2007L, 2008L, 2009L, 2010L, 2005L, 
2006L, 2007L, 2008L, 2009L, 2010L, 2005L, 2006L, 2007L, 2008L, 
2009L, 2010L, 2005L, 2006L, 2007L, 2008L, 2009L, 2010L), cab = c(28.2321949, 
29.25037956, 37.5, 31.28475761, 32.32411957, 33.39371109, 26.71179771, 
27.9, 28.71139336, 29.71598816, 30.73922729, 34.2, 44.23, 50.52510246, 
46.5, 52.65625763, 55.14787292, 55.26, 58.9, 53.74567795, 54.78623962, 
55.83730316, 56.8971138, 57.96392822), dit = c(13.89709218, 20.40781491, 
16.2123884, 8.125547549, -10.76938871, 8.957039766, -3.445136313, 
-3.090532362, -6.554877979, -5.520878675, -6.717259628, -5.567234574, 
-0.250193221, 1.66538338, 1.076211925, 1.01077433, 3.470143475, 
1.82904182, 0.232220267, 2.690679668, -0.683459128, 1.864047049, 
6.662914109, 1.866754256), fdi = c(-3.526655479, -0.072001021, 
-1.368761628, 1.896315051, 3.136662133, -3.851110464, -0.133873602, 
-0.175920935, 1.706123419, 0.49455672, -0.193899058, 0.561144791, 
1.17120896, 0.635657188, 0.817754424, 1.449748396, 0.879494542, 
1.068934861, 2.042272555, 1.969861557, 2.178644539, 1.925660984, 
1.631267003, 1.439124272), government = c(16.02903718, 15.34172241, 
15.5369347, 16.81461146, 19.89859034, 17.04234392, 8.991073644, 
9.666734707, 9.568485256, 10.23358845, 11.28862086, 11.21186949, 
5.180225165, 5.440078557, 5.359456215, 5.178276878, 5.093745166, 
5.075325836, 14.42744504, 13.59161479, 13.33857571, 12.66976339, 
12.63534697, 12.17758809)), class = "data.frame", row.names = c(NA, 
-24L))

Answer 1

You need to exclude the "time" column from your gather call, since you need each row to have a value for time.

You also need to supply an x and y aesthetic to your geom_line call.

Try this:

exampl%>%
  tidyr::gather(key = "key", value = "value", -time, -unit_id) %>%
  ggplot(aes(x = time, y = value, color = unit_id)) +       
  facet_wrap(~ key, scales = "free") +   # In separate panels
  geom_line()

You may also want to look into pivot_longer() . It is a new version of gather that might be a little more user friendly.

Edit (thanks r2evans), here it is with pivot_longer.

exampl%>%
  tidyr::pivot_longer(
    -c(time, unit_id), 
    names_to = "key", 
    values_to = "value"
  ) %>%
  ggplot(aes(x = time, y = value, color = unit_id)) +                     
  facet_wrap(~ key, scales = "free") +   
  geom_line()