Donut VPC works for complete dataset, not for subset

[chenxuli-umn]

Hello,

Nice work for this package.

Recently, I'm trying to use this for my own project. However, I encountered some issues when plotting the donut VPC.

It works for my complete dataset:
``
df2 <- df %>% filter(TIME == 0) %>% select(WT, Alb, CRP)

cop <- vine(df2)

sim <- rcopula(cop, sim_nr = 100)

donut <- donutVPC( sim_data = sim, obs_data = df2, percentiles = c(5, 50, 95),
sim_nr = 100, conf_band = 99,
colors_bands = c("#99E0DC", "#E498B4"),
cores = 4)

``

But it doesn’t work for the subset (PER refers to periods of a individual, study is cross over designed with 2 periods per individual) :
``
df3 <- df %>% filter(TIME == 0, PER ==1) %>% select(WT, Alb, CRP)

cop <- vine(df3)

sim <- rcopula(cop, sim_nr = 100)

donut <- donutVPC( sim_data = sim, obs_data = df3, percentiles = c(5, 50, 95),
sim_nr = 100, conf_band = 99,
colors_bands = c("#99E0DC", "#E498B4"),
cores = 4)

``

It gives me an error message shown below:
``
Error in dplyr::mutate():
ℹ In argument: `percentile =
names(contour_levels[contour_levels == level[1]])`.
Caused by error:
! `percentile` must be size 139 or 1, not 2.
Run `rlang::last_trace()` to see where the error occurred.

``

Appreciate any insights on this issue.

[laurazwep]

Are you still experiencing this issue? From your code I cannot tell any problems. How large is your dataset df3 (number of rows)? We will have a look at the internal code in the coming month and hopefully will solve this problem