bootstrap_density_tbl

[spsanderson]

Function is not yet fully outlined but it is something like this:

library(TidyDensity)
library(dplyr)
library(purrr)
library(tidyr)
library(ggplot2)

x <- mtcars$mpg

# Add density tibble from stats::dens()
bootstrap_density_tbl <- tidy_bootstrap(x) %>%
  bootstrap_unnest_tbl() %>%
  nest_by(sim_number) %>%
  mutate(dens_tbl = list(density(unlist(data), n = nrow(data))[c("x","y")] %>%
                           set_names("dx","dy") %>%
                           as_tibble())) %>%
  select(sim_number, dens_tbl) %>%
  unnest(cols = dens_tbl) %>%
  mutate(x = row_number()) %>%
  ungroup()

> bootstrap_density_tbl
# A tibble: 50,000 × 4
   sim_number    dx       dy     x
   <fct>      <dbl>    <dbl> <int>
 1 1           6.06 0.000124     1
 2 1           7.40 0.00130      2
 3 1           8.74 0.00581      3
 4 1          10.1  0.0118       4
 5 1          11.4  0.0143       5
 6 1          12.8  0.0204       6
 7 1          14.1  0.0360       7
 8 1          15.4  0.0582       8
 9 1          16.8  0.0826       9
10 1          18.1  0.104       10

bootstrap_density_tbl %>%
  ggplot(aes(x = dx, y = dy, group = sim_number)) +
  geom_line(color = "gray") +
  geom_smooth(group = "none", se = FALSE, color = "red", linetype = "dashed") +
  theme_minimal()

bootstrap_density_tbl %>%
  group_by(x) %>%
  summarise(
    dx = mean(dx),
    dy = mean(dy)
  ) %>%
  ungroup() %>%
  ggplot(aes(x = dx, y = dy)) +
  geom_line(
    data = bootstrap_density_tbl %>%
      select(dx, dy, sim_number),
    aes(x = dx, y = dy, group = sim_number),
    color = "gray"
  ) + 
  geom_line() +
  theme_minimal()

[spsanderson]

Possible function:

bootstrap_density_augment <- function(.data){
  
  atb <- attributes(.data)
  
  # Checks 
  if (!is.data.frame(.data)){
    rlang::abort(
      message = "'.data' is expecting a data.frame/tibble. Please supply.",
      use_cli_format = TRUE
    )
  }
  
  if (!atb$tibble_type %in% c("tidy_bootstrap","tidy_bootstrap_nested")){
    rlang::abort(
      message = "Must pass data to this function from either tidy_bootstrap() or 
      bootstrap_unnest_tbl().",
      use_cli_format = TRUE
    )
  }
  
  # Add density data
  if(atb$tibble_type == "tidy_bootstrap_nested"){
    df_tbl <- dplyr::as_tibble(.data) %>% 
      TidyDensity::bootstrap_unnest_tbl()
  }
  
  if(atb$tibble_type == "tidy_bootstrap"){
    df_tbl <- dplyr::as_tibble(.data)
  }
  
  df_tbl <- df_tbl %>%
    dplyr::nest_by(sim_number) %>%
    dplyr::mutate(dens_tbl = list(
      stats::density(unlist(data),
                     n = nrow(data))[c("x","y")] %>%
        purrr::set_names("dx","dy") %>%
        dplyr::as_tibble())) %>%
    tidyr::unnest(cols = c(data, dens_tbl)) %>%
    dplyr::mutate(x = dplyr::row_number()) %>%
    dplyr::ungroup() %>%
    dplyr::select(sim_number, x, y, dx, dy)

  # Return
  return(df_tbl)

}

Example:

library(TidyDensity)
library(tidyverse)

x <- mtcars$mpg

# Add density tibble from stats::dens()
bootstrap_density_tbl <- tidy_bootstrap(x)

> bootstrap_density_augment(bootstrap_density_tbl)
# A tibble: 50,000 × 5
   sim_number     x     y    dx        dy
   <fct>      <int> <dbl> <dbl>     <dbl>
 1 1              1  19.7  3.73 0.0000824
 2 1              2  27.3  5.27 0.000529 
 3 1              3  10.4  6.80 0.00223  
 4 1              4  15.5  8.34 0.00663  
 5 1              5  16.4  9.87 0.0159   
 6 1              6  13.3 11.4  0.0336   
 7 1              7  14.7 12.9  0.0597   
 8 1              8  26   14.5  0.0808   
 9 1              9  14.7 16.0  0.0812   
10 1             10  21   17.5  0.0670   
# … with 49,990 more rows
# ℹ Use `print(n = ...)` to see more rows

OR with table unnested already:

bootstrap_density_tbl <- tidy_bootstrap(x) %>%
  bootstrap_unnest_tbl()

> bootstrap_density_augment(bootstrap_density_tbl)
# A tibble: 50,000 × 5
   sim_number     x     y    dx       dy
   <fct>      <int> <dbl> <dbl>    <dbl>
 1 1              1  26    4.77 0.000106
 2 1              2  15    6.34 0.000563
 3 1              3  17.3  7.91 0.00229 
 4 1              4  24.4  9.48 0.00718 
 5 1              5  15.8 11.0  0.0175  
 6 1              6  24.4 12.6  0.0335  
 7 1              7  30.4 14.2  0.0505  
 8 1              8  17.3 15.8  0.0616  
 9 1              9  22.8 17.3  0.0633  
10 1             10  33.9 18.9  0.0587  
# … with 49,990 more rows
# ℹ Use `print(n = ...)` to see more rows