3

I tried to create highchart density with more than two groups. I found a way to add them one by one manually but there must be a better way to handle groups.

Examples: I would like to create a highchart similar to ggplot chart below without adding them one by one. Is there any way to do so?

d

f <- data.frame(MEI = c(-2.031, -1.999, -1.945, -1.944, -1.875, 
                       -1.873, -1.846, -2.031, -1.999, -1.945, -1.944, -1.875, -1.873, 
                       -1.846, -2.031, -1.999, -1.945, -1.944, -1.875, -1.873, -1.846, 
                       -2.031, -1.999, -1.945, -1.944, -1.875, -1.873, -1.846), 
                 Count = c(10L,0L, 15L, 1L, 6L, 10L, 18L, 10L, 0L, 15L, 1L, 6L, 10L, 0L, 15L, 
                          10L, 0L, 15L, 1L, 6L, 10L, 10L, 0L, 15L, 1L, 6L, 10L, 18L), 
                 Region = c("MidWest", "MidWest", "MidWest", "MidWest", "MidWest", "MidWest", "MidWest", 
                                        "South", "South", "South", "South", "South", "South", "South", 
                                        "South", "South", "South", "NorthEast", "NorthEast", "NorthEast", 
                                        "NorthEast", "NorthEast", "NorthEast", "NorthEast", "NorthEast", 
                                        "NorthEast", "NorthEast", "NorthEast"))
df <- data.table(ddf)
df %>%ggplot() + 
  geom_density(aes(x=MEI, group=Region, fill=Region),alpha=0.5) + 
  xlab("MEI") +
  ylab("Density")

hcdensity(df[Region=="NorthEast"]$MEI,area = TRUE) %>%
  hc_add_series(density(df[Region=="MidWest"]$MEI), area = TRUE) %>%
  hc_add_series(density(df[Region=="South"]$MEI), area = TRUE)
southwind
  • 636
  • 4
  • 15

2 Answers2

1

I couldn't figure out a way to do so directly in highcharter. But if you calculate the densities beforehand and use purrr's reduce() function you can automatize the plot creation:

library(purrr)

# calculate a list of densities (one per region)
densities <- df %>% group_by(Region) %>% 
  do(den = density(.$MEI)) %>%
  .$den

# create the highchart with all densities
reduce(densities, hc_add_series, .init = highchart())

reduce() combines the list into a single object. To the first density in the list highchart() is applied and thus the necessary highchart htmlwidget is created. All the other densities are then added to it with hc_add_series().

wici
  • 1,681
  • 1
  • 14
  • 21
1

Method 1: tapply + reduce + hc_add_series

tapply(df$MEI, df$Region, density) %>%
  reduce(.f = hc_add_series, .init = highchart())

Method 2: map + hc_add_series_list

(Reference: RPubs - Highcharter hc_add_series_list)

ds <- map(levels(df$Region), function(x){
  dt <- density(df$MEI[df$Region == x])[1:2]
  dt <- list_parse2(as.data.frame(dt))
  list(data = dt, name = x)
})

highchart() %>% 
  hc_add_series_list(ds)

enter image description here

Darren Tsai
  • 32,117
  • 5
  • 21
  • 51