Closed spsanderson closed 2 years ago
Function:
#' Distribution Statistics
#'
#' @family Binomial
#' @fmaily Distribution Statistics
#'
#' @author Steven P. Sanderson II, MPH
#'
#' @details This function will take in a tibble and returns the statistics
#' of the given type of `tidy_` distribution. It is required that data be
#' passed from a `tidy_` distribution function.
#'
#' @description Returns distribution statistics in a tibble.
#'
#' @param .data The data being passed from a `tidy_` distribution function.
#'
#' @examples
#' tidy_binomial() %>%
#' util_binomial_stats_tbl()
#'
#' @return
#' A tibble
#'
#' @export
#'
util_binomial_stats_tbl <- function(.data){
# Immediate check for tidy_ distribution function
if (!"tibble_type" %in% names(attributes(.data))){
rlang::abort(
message = "You must pass data from the 'tidy_dist' function.",
use_cli_format = TRUE
)
}
if (attributes(.data)$tibble_type != "tidy_binomial"){
rlang::abort(
message = "You must use 'tidy_binomial()'",
use_cli_format = TRUE
)
}
# Data
data_tbl <- tibble::as_tibble(.data)
atb <- attributes(data_tbl)
n <- atb$.size
p <- atb$.prob
stat_mean <- n*p
stat_mode <- c(p * (n + 1) - 1, p * (n + 1))
stat_sd <- sqrt( (p*q)/((p+q)^2 * (p + q + 1)) )
stat_skewness <- (1 - 2*p)/sqrt((n*p) * (1-p))
stat_kurtosis <- 3 - 6/n + 1/((n*p) * (1 - p))
stat_coef_var <- sqrt((1-p)/(n*p))
# Data Tibble
ret <- tibble::tibble(
tidy_function = atb$tibble_type,
function_call = atb$dist_with_params,
distribution = atb$tibble_type %>%
stringr::str_remove("tidy_") %>%
stringr::str_to_title(),
distribution_type = atb$distribution_family_type,
points = atb$.n,
simulations = atb$.num_sims,
mean = stat_mean,
mode_lower = stat_mode[[1]],
mode_upper = stat_mode[[2]],
range = paste0("0 to ", n),
std_dv = stat_sd,
coeff_var = stat_coef_var,
skewness = stat_skewness,
kurtosis = stat_kurtosis,
computed_std_skew = tidy_skewness_vec(data_tbl$y),
computed_std_kurt = tidy_kurtosis_vec(data_tbl$y)
)
# Return
return(ret)
}
Example:
tidy_binomial(.size = 2, .prob = 0.25) %>%
util_binomial_stats_tbl() %>%
glimpse()
Rows: 1
Columns: 16
$ tidy_function <chr> "tidy_binomial"
$ function_call <chr> "Binomial c(2, 0.25)"
$ distribution <chr> "Binomial"
$ distribution_type <chr> "discrete"
$ points <dbl> 50
$ simulations <dbl> 1
$ mean <dbl> 0.5
$ mode_lower <dbl> -0.25
$ mode_upper <dbl> 0.75
$ range <chr> "0 to 2"
$ std_dv <dbl> 0.2666667
$ coeff_var <dbl> 1.224745
$ skewness <dbl> 0.8164966
$ kurtosis <dbl> 2.666667
$ computed_std_skew <dbl> 0.5466228
$ computed_std_kurt <dbl> 2.224713
https://www.itl.nist.gov/div898/handbook/eda/section3/eda366i.htm