Fixes #189

DESCRIPTION

@@ -13,7 +13,7 @@ Description:
 License: MIT + file LICENSE
 Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.1.2
+RoxygenNote: 7.2.0
 URL: https://github.com/spsanderson/TidyDensity
 BugReports: https://github.com/spsanderson/TidyDensity/issues
 Imports: 

NAMESPACE

@@ -15,6 +15,7 @@ export(td_scale_fill_colorblind)
 export(tidy_autoplot)
 export(tidy_beta)
 export(tidy_binomial)
+export(tidy_bootstrap)
 export(tidy_burr)
 export(tidy_cauchy)
 export(tidy_chisquare)

NEWS.md

@@ -7,6 +7,7 @@ None
 1. Fix #181 - Add functions `color_blind()` `td_scale_fill_colorblind()` and 
 `td_scale_color_colorblind()`
 2. Fix #187 - Add functions `ci_lo()` and `ci_hi()`
+3. Fix #189 - Add function `tidy_bootstrap()`
 
 ## Minor Fixes and Improvements
 1. Fix #176 - Update `_autoplot` functions to include cumulative mean MCMC chart

R/empirical-tidy-bootstrap.R

@@ -0,0 +1,95 @@
+#' Bootstrap Empirical Data
+#'
+#' @family Bootstrap
+#'
+#' @author Steven P. Sanderson II, MPH
+#'
+#' @details This function will take in a numeric input vector and produce a tibble
+#' of bootstrapped values in a list. The table that is output will have two columns:
+#' `sim_number` and `bootstrap_samples`
+#'
+#' The `sim_number` corresponds to how many times you want the data to be resampled,
+#' and the `bootstrap_samples` column contains a list of the boostrapped resampled
+#' data.
+#'
+#' @description Takes an input vector of numeric data and produces a bootstrapped
+#' nested tibble by simulation number.
+#'
+#' @param .x The vector of data being passed to the function. Must be a numeric
+#' vector.
+#' @param .num_sims The default is 2000, can be set to anything desired. A warning
+#' will pass to the console if the value is less than 2000.
+#' @param .proportion How much of the original data do you want to pass through
+#' to the sampling function. The default is 0.80 (80%)
+#' @param .distribution_type This can either be 'continuous' or 'discrete'
+#'
+#' @examples
+#' x <- mtcars$mpg
+#' tidy_bootstrap(x)
+#'
+#' @return
+#' A nested tibble
+#'
+#' @export
+#'
+
+tidy_bootstrap <- function(.x, .num_sims = 2000, .proportion = 0.8,
+                           .distribution_type = "continuous"){
+
+    # Tidyeval ----
+    x_term <- as.numeric(.x)
+    n <- length(x_term)
+    dist_type <- tolower(as.character(.distribution_type))
+    num_sims <- as.integer(.num_sims)
+    size <- as.numeric(.proportion)
+
+    # Checks ----
+    if (!is.vector(x_term)) {
+        rlang::abort(
+            message = "You must pass a vector as the .x argument to this function.",
+            use_cli_format = TRUE
+        )
+    }
+
+    if (size < 0 | size > 1){
+        rlang::abort(
+            message = "The '.proportion' parameter must be between 0 and 1 inclusive.",
+            use_cli_format = TRUE
+        )
+    }
+
+    if (!dist_type %in% c("continuous","discrete")){
+        rlang::abort(
+            message = "You must choose either 'continuous' or 'discrete'.",
+            use_cli_format = TRUE
+        )
+    }
+
+    if (num_sims < 2000){
+        rlang::warn(
+            message = "Setting '.num_sims' to less than 2000 means that results can be
+      potentially unstable. Consider setting to 2000 or more.",
+            use_cli_format = TRUE
+        )
+    }
+
+    # Data ----
+    df <- dplyr::tibble(sim_number = as.factor(1:num_sims)) %>%
+        dplyr::group_by(sim_number) %>%
+        dplyr::mutate(bootstrap_samples = list(
+            sample(x = x_term, size = floor(size * n) ,replace = TRUE)
+          )
+        ) %>%
+        dplyr::ungroup()
+
+    # Attach descriptive attributes to tibble
+    attr(df, "distribution_family_type") <- dist_type
+    attr(df, ".x") <- .x
+    attr(df, ".num_sims") <- .num_sims
+    attr(df, "tibble_type") <- "tidy_bootstrap"
+    attr(df, "dist_with_params") <- "Empirical"
+
+    # Return ----
+    return(df)
+
+}

_pkgdown.yml

@@ -37,6 +37,10 @@ reference:
      desc: Function for creating mixture model data
      contents:
      -  has_concept("Mixture Data")
+  -  title: Bootstrap
+     desc: Functions for bootstrapping data
+     contents:
+     -  has_concept("Bootstrap")
   -  title: Parameter Estimation Functions
      desc: Functions that help to estimate parameters from raw data.
      contents: