Visualization of Groups of Poisson random samples using ggridges - r

Question

I have two sets of data, all in one data frame. The first set is related to data collected in Location 1 and the second set is collected in Location 2. Each location has different count data (column value) for 5 months.
# DataSet
-----------------
rp_data <- structure(list(Month = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L,
3L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L,
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 4L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L), .Label = c("1",
"2", "3", "4", "5"), class = "factor"), location = c("1", "1",
"1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1",
"1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1",
"1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1",
"1", "1", "1", "1", "1", "1", "1", "1", "1", "2", "2", "2", "2",
"2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2",
"2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2",
"2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2",
"2", "2", "2", "2", "2", "2", "2"), value = c(0L, 1L, 1L, 1L,
2L, 1L, 0L, 0L, 1L, 1L, 3L, 2L, 1L, 4L, 1L, 3L, 1L, 1L, 1L, 1L,
2L, 2L, 1L, 0L, 2L, 4L, 3L, 5L, 5L, 0L, 4L, 3L, 3L, 4L, 2L, 5L,
2L, 3L, 10L, 6L, 5L, 6L, 4L, 6L, 4L, 5L, 6L, 5L, 3L, 7L, 1L,
1L, 1L, 1L, 0L, 0L, 2L, 1L, 2L, 0L, 2L, 3L, 4L, 1L, 2L, 1L, 2L,
0L, 2L, 2L, 4L, 4L, 5L, 1L, 4L, 5L, 4L, 5L, 1L, 4L, 3L, 7L, 7L,
4L, 2L, 5L, 4L, 1L, 5L, 3L, 7L, 3L, 4L, 8L, 5L, 7L, 1L, 1L, 6L,
3L)), .Names = c("Month", "location", "value"), row.names = c(NA,
-100L), class = "data.frame")
I used this example below, as illustrated on the ggridges examples webpage, to display the various count values across different months.
# Plot 1 , filtering data related to location = 1
#---------------
ggplot(rp_data[rp_data$location == '1',], aes(x = value, y = Month, group = Month)) +
geom_density_ridges2(aes(fill = Month), stat = "binline", binwidth = 1, scale = 0.95) +
geom_text(stat = "bin",
aes(y = group + 0.95*(..count../max(..count..)),
label = ifelse(..count..>0, ..count.., "")),
vjust = 1.4, size = 3, color = "white", binwidth = 1) +
scale_x_continuous(breaks = c(0:12), limits = c(-.5, 13), expand = c(0, 0),
name = "random value") +
scale_y_discrete(expand = c(0.01, 0), name = "Month",
labels = c("5.0", "4.0", "3.0", "2.0", "1.0")) +
scale_fill_cyclical(values = c("#0000B0", "#7070D0")) +
labs(title = "Poisson random samples location 1 different Month",
subtitle = "sample size n=10") +
guides(y = "none") +
theme_ridges(grid = FALSE) +
theme(axis.title.x = element_text(hjust = 0.5),
axis.title.y = element_text(hjust = 0.5))
# Plot 2 , filtering data related to location = 2
#---------------
ggplot(rp_data[rp_data$location == '2',], aes(x = value, y = Month, group = Month)) +
geom_density_ridges2(aes(fill = Month), stat = "binline", binwidth = 1, scale = 0.95) +
geom_text(stat = "bin",
aes(y = group + 0.95*(..count../max(..count..)),
label = ifelse(..count..>0, ..count.., "")),
vjust = 1.4, size = 3, color = "white", binwidth = 1) +
scale_x_continuous(breaks = c(0:12), limits = c(-.5, 13), expand = c(0, 0),
name = "random value") +
scale_y_discrete(expand = c(0.01, 0), name = "Month",
labels = c("5.0", "4.0", "3.0", "2.0", "1.0")) +
scale_fill_cyclical(values = c("#0000B0", "#7070D0")) +
labs(title = "Poisson random samples location 2 different Month",
subtitle = "sample size n=10") +
guides(y = "none") +
theme_ridges(grid = FALSE) +
theme(axis.title.x = element_text(hjust = 0.5),
axis.title.y = element_text(hjust = 0.5))
Result for plot 1:
My question is how can I combine these two plots, sort of like an overlay plot as shown in this example:
I don't want to plot them in two separate plots.

Answer 1

You need to create a grouping variable that contains both Month and location. You can do that by using paste0(Month, location). For now, I'm leaving out the text labels, though they may be possible with a little more thought as well. (But I think they'd make the figure too busy.)
ggplot(rp_data,
aes(x = value, y = Month,
group = paste0(Month, location),
fill = paste0(Month, location))) +
geom_density_ridges2(stat = "binline", binwidth = 1,
scale = 0.95, alpha = 0.7) +
scale_x_continuous(breaks = c(0:12), limits = c(-.5, 13),
expand = c(0, 0), name = "random value") +
scale_y_discrete(expand = c(0.01, 0), name = "Month",
labels = c("5.0", "4.0", "3.0", "2.0", "1.0")) +
scale_fill_cyclical(values = c("#0000B0", "#B00000",
"#7070D0", "#FC5E5E")) +
labs(title = "Poisson random samples location 1 different Month",
subtitle = "sample size n=10") +
guides(y = "none") +
theme_ridges(grid = FALSE, center = TRUE)
Edit: Now with text labels.
ggplot(rp_data, aes(x = value, y = Month, group = paste0(Month, location), fill = paste0(Month, location))) +
geom_density_ridges2(stat = "binline", binwidth = 1, scale = 0.95, alpha = 0.7) +
geom_text(stat = "bin",
aes(y = ceiling(group/2) + 0.95*(..count../max(..count..)),
label = ifelse(..count..>0, ..count.., ""), color = location),
vjust = 1.4, size = 3, binwidth = 1, fontface = "bold") +
scale_x_continuous(breaks = c(0:12), limits = c(-.5, 13), expand = c(0, 0),
name = "random value") +
scale_y_discrete(expand = c(0.01, 0), name = "Month",
labels = c("5.0", "4.0", "3.0", "2.0", "1.0")) +
scale_fill_cyclical(values = c("#0000B0", "#B00000", "#7070D0", "#FC5E5E")) +
scale_color_cyclical(values = c("white", "black")) +
labs(title = "Poisson random samples location 1 different Month",
subtitle = "sample size n=10") +
guides(y = "none") +
theme_ridges(grid = FALSE, center = TRUE)
Again, not sure it's a good idea, but there you go.