Merge pull request #497 from spsanderson/development

Fixes #477

NAMESPACE

@@ -117,6 +117,9 @@ export(util_geometric_stats_tbl)
 export(util_hypergeometric_aic)
 export(util_hypergeometric_param_estimate)
 export(util_hypergeometric_stats_tbl)
+export(util_inverse_weibull_aic)
+export(util_inverse_weibull_param_estimate)
+export(util_inverse_weibull_stats_tbl)
 export(util_logistic_aic)
 export(util_logistic_param_estimate)
 export(util_logistic_stats_tbl)

NEWS.md

@@ -27,6 +27,9 @@ Add function `util_pareto1_stats_tbl()` to create a summary table of the Pareto
 9. Fix #478 - Add function `util_paralogistic_param_estimate()` to estimate the parameters of the paralogistic distribution.
 Add function `util_paralogistic_aic()` to calculate the AIC for the paralogistic distribution.
 Add fnction `util_paralogistic_stats_tbl()` to create a summary table of the paralogistic distribution.
+10. Fix #477 - Add function `util_inverse_weibull_param_estimate()` to estimate the parameters of the Inverse Weibull distribution.
+Add function `util_inverse_weibull_aic()` to calculate the AIC for the Inverse Weibull distribution.
+Add function `util_inverse_weibull_stats_tbl()` to create a summary table of the Inverse Weibull distribution.
 
 ## Minor Improvements and Fixes
 1. Fix #468 - Update `util_negative_binomial_param_estimate()` to add the use of

R/est-param-inv-weibull.R

@@ -0,0 +1,119 @@
+#' Estimate Inverse Weibull Parameters
+#'
+#' @family Parameter Estimation
+#' @family Inverse Weibull
+#'
+#' @author Steven P. Sanderson II, MPH
+#'
+#' @details This function will attempt to estimate the inverse Weibull shape and rate
+#' parameters given some vector of values.
+#'
+#' @description The function will return a list output by default, and if the parameter
+#' `.auto_gen_empirical` is set to `TRUE` then the empirical data given to the
+#' parameter `.x` will be run through the `tidy_empirical()` function and combined
+#' with the estimated inverse Weibull data.
+#'
+#' @param .x The vector of data to be passed to the function.
+#' @param .auto_gen_empirical This is a boolean value of TRUE/FALSE with default
+#' set to TRUE. This will automatically create the `tidy_empirical()` output
+#' for the `.x` parameter and use the `tidy_combine_distributions()`. The user
+#' can then plot out the data using `$combined_data_tbl` from the function output.
+#'
+#' @examples
+#' library(dplyr)
+#' library(ggplot2)
+#'
+#' set.seed(123)
+#' x <- tidy_inverse_weibull(100, .shape = 2, .scale = 1)[["y"]]
+#' output <- util_inverse_weibull_param_estimate(x)
+#'
+#' output$parameter_tbl
+#'
+#' output$combined_data_tbl %>%
+#'   tidy_combined_autoplot()
+#'
+#' @return
+#' A tibble/list
+#'
+#' @name util_inverse_weibull_param_estimate
+NULL
+#' @export
+#' @rdname util_inverse_weibull_param_estimate
+
+util_inverse_weibull_param_estimate <- function(.x, .auto_gen_empirical = TRUE) {
+
+  # Tidyeval ----
+  x_term <- as.numeric(.x)
+  minx <- min(x_term)
+  maxx <- max(x_term)
+  n <- length(x_term)
+  unique_terms <- length(unique(x_term))
+
+  # Checks ----
+  if (!is.numeric(.x)) {
+    rlang::abort(
+      message = "The '.x' parameter must be a numeric vector.",
+      use_cli_format = TRUE
+    )
+  }
+
+  # Negative log-likelihood function ----
+  neg_log_lik <- function(params, data) {
+    shape <- params[1]
+    scale <- params[2]
+    -sum(actuar::dinvweibull(data, shape = shape, scale = scale, log = TRUE))
+  }
+
+  # Initial parameter guesses
+  initial_params <- c(shape = 1, scale = 1)
+
+  # Optimize to minimize the negative log-likelihood
+  opt_result <- optim(
+    par = initial_params,
+    fn = neg_log_lik,
+    data = x_term,
+    method = "L-BFGS-B",
+    lower = c(1e-5, 1e-5)
+  )
+
+  iw_shape <- opt_result$par[1]
+  iw_scale <- opt_result$par[2]
+  iw_rate <- 1 / iw_scale
+
+  # Return Tibble ----
+  if (.auto_gen_empirical) {
+    te <- tidy_empirical(.x = x_term)
+    td <- tidy_inverse_weibull(.n = n, .shape = round(iw_shape, 3), .rate = round(iw_rate, 3))
+    combined_tbl <- tidy_combine_distributions(te, td)
+  }
+
+  ret <- dplyr::tibble(
+    dist_type = "Inverse Weibull",
+    samp_size = n,
+    min = minx,
+    max = maxx,
+    method = "MLE",
+    shape = iw_shape,
+    scale = iw_scale,
+    rate = iw_rate
+  )
+
+  # Return ----
+  attr(ret, "tibble_type") <- "parameter_estimation"
+  attr(ret, "family") <- "inverse_weibull"
+  attr(ret, "x_term") <- .x
+  attr(ret, "n") <- n
+
+  if (.auto_gen_empirical) {
+    output <- list(
+      combined_data_tbl = combined_tbl,
+      parameter_tbl     = ret
+    )
+  } else {
+    output <- list(
+      parameter_tbl = ret
+    )
+  }
+
+  return(output)
+}

R/stats-inv-weibull-tbl.R

@@ -0,0 +1,108 @@
+#' Distribution Statistics
+#'
+#' @family Inverse Weibull
+#' @family 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
+#' library(dplyr)
+#'
+#' set.seed(123)
+#' tidy_inverse_weibull() |>
+#'   util_inverse_weibull_stats_tbl() |>
+#'   glimpse()
+#'
+#' @return
+#' A tibble
+#'
+#' @name util_inverse_weibull_stats_tbl
+NULL
+#' @export
+#' @rdname util_inverse_weibull_stats_tbl
+
+util_inverse_weibull_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)$tibble_type != "tidy_inverse_weibull") {
+    rlang::abort(
+      message = "You must use 'tidy_inverse_weibull()'",
+      use_cli_format = TRUE
+    )
+  }
+
+  # Data
+  data_tbl <- dplyr::as_tibble(.data)
+
+  atb <- attributes(data_tbl)
+  t <- atb$.shape
+  q <- atb$.scale
+
+  # Negative log-likelihood function ----
+  neg_log_lik <- function(params, data) {
+    shape <- params[1]
+    scale <- params[2]
+    -sum(actuar::dinvweibull(data, shape = shape, scale = scale, log = TRUE))
+  }
+
+  # Initial parameter guesses
+  initial_params <- c(shape = t, scale = q)
+
+  # Optimize to minimize the negative log-likelihood
+  opt_result <- optim(
+    par = initial_params,
+    fn = neg_log_lik,
+    data = data_tbl$y,
+    method = "L-BFGS-B",
+    lower = c(1e-5, 1e-5)
+  )
+
+  iw_shape <- opt_result$par[1]
+  iw_scale <- opt_result$par[2]
+  iw_rate <- 1 / iw_scale
+
+  # Compute statistics
+  stat_mean <- mean(actuar::rinvweibull(1e5, shape = iw_shape, scale = iw_scale))
+  stat_median <- quantile(data_tbl$y, 0.5)
+  stat_mode <- iw_scale * (1 - 1 / iw_shape)^(1 / iw_shape)
+  stat_sd <- sqrt(var(actuar::rinvweibull(1e5, shape = iw_shape, scale = iw_scale)))
+  stat_coef_var <- stat_sd / stat_mean
+
+  # Data Tibble
+  ret <- dplyr::tibble(
+    tidy_function = atb$tibble_type,
+    function_call = atb$dist_with_params,
+    distribution = dist_type_extractor(atb$tibble_type),
+    distribution_type = atb$distribution_family_type,
+    points = atb$.n,
+    simulations = atb$.num_sims,
+    mean = stat_mean,
+    median = stat_median,
+    mode = stat_mode,
+    range = paste0("0 to Inf"),
+    std_dv = stat_sd,
+    coeff_var = stat_coef_var,
+    computed_std_skew = tidy_skewness_vec(data_tbl$y),
+    computed_std_kurt = tidy_kurtosis_vec(data_tbl$y),
+    ci_lo = ci_lo(data_tbl$y),
+    ci_hi = ci_hi(data_tbl$y)
+  )
+
+  # Return
+  return(ret)
+}

R/utils-aic-inv-weibull.R

@@ -0,0 +1,81 @@
+#' Calculate Akaike Information Criterion (AIC) for Inverse Weibull Distribution
+#'
+#' This function calculates the Akaike Information Criterion (AIC) for an inverse Weibull
+#' distribution fitted to the provided data.
+#'
+#' @family Utility
+#'
+#' @author Steven P. Sanderson II, MPH
+#'
+#' @description
+#' This function estimates the shape and scale parameters of an inverse Weibull distribution
+#' from the provided data using maximum likelihood estimation,
+#' and then calculates the AIC value based on the fitted distribution.
+#'
+#' @param .x A numeric vector containing the data to be fitted to an inverse Weibull distribution.
+#'
+#' @details
+#' This function fits an inverse Weibull distribution to the provided data using maximum
+#' likelihood estimation. It estimates the shape and scale parameters
+#' of the inverse Weibull distribution using maximum likelihood estimation. Then, it
+#' calculates the AIC value based on the fitted distribution.
+#'
+#' Initial parameter estimates: The function uses the method of moments estimates
+#' as starting points for the shape and scale parameters of the inverse Weibull
+#' distribution.
+#'
+#' Optimization method: The function uses the optim function for optimization.
+#' You might explore different optimization methods within optim for potentially
+#' better performance.
+#'
+#' Goodness-of-fit: While AIC is a useful metric for model comparison, it's recommended
+#' to also assess the goodness-of-fit of the chosen model using visualization
+#' and other statistical tests.
+#'
+#' @examples
+#' # Example 1: Calculate AIC for a sample dataset
+#' set.seed(123)
+#' x <- tidy_inverse_weibull(.n = 100, .shape = 2, .scale = 1)[["y"]]
+#' util_inverse_weibull_aic(x)
+#'
+#' @return
+#' The AIC value calculated based on the fitted inverse Weibull distribution to the provided data.
+#'
+#' @name util_inverse_weibull_aic
+NULL
+#' @export
+#' @rdname util_inverse_weibull_aic
+util_inverse_weibull_aic <- function(.x) {
+  # Tidyeval
+  x <- as.numeric(.x)
+
+  # Negative log-likelihood function for inverse Weibull distribution
+  neg_log_lik_invweibull <- function(par, data) {
+    shape <- par[1]
+    scale <- par[2]
+    -sum(actuar::dinvweibull(data, shape = shape, scale = scale, log = TRUE))
+  }
+
+  # Get initial parameter estimates: method of moments
+  # Note: This assumes the availability of a suitable method for initial parameter estimation.
+  initial_params <- c(shape = 1, scale = 1)
+
+  # Fit inverse Weibull distribution using optim
+  fit_invweibull <- optim(
+    par = initial_params,
+    fn = neg_log_lik_invweibull,
+    data = x,
+    method = "L-BFGS-B",
+    lower = c(1e-5, 1e-5)
+  )
+
+  # Extract log-likelihood and number of parameters
+  logLik_invweibull <- -fit_invweibull$value
+  k_invweibull <- 2 # Number of parameters for inverse Weibull distribution (shape and scale)
+
+  # Calculate AIC
+  AIC_invweibull <- 2 * k_invweibull - 2 * logLik_invweibull
+
+  # Return AIC
+  return(AIC_invweibull)
+}

docs/pkgdown.yml

@@ -3,7 +3,7 @@ pkgdown: 2.0.9
 pkgdown_sha: ~
 articles:
   getting-started: getting-started.html
-last_built: 2024-05-15T01:57Z
+last_built: 2024-05-15T15:39Z
 urls:
   reference: https://www.spsanderson.com/TidyDensity/reference
   article: https://www.spsanderson.com/TidyDensity/articles