Merge pull request #451 from spsanderson/development

Fixes #434
spsanderson · Apr 25, 2024 · e225102 · e225102
2 parents cbf00ff + d686898
commit e225102
Show file tree

Hide file tree

Showing 19 changed files with 143 additions and 0 deletions.
diff --git a/NAMESPACE b/NAMESPACE
@@ -109,6 +109,7 @@ export(util_f_stats_tbl)
 export(util_gamma_aic)
 export(util_gamma_param_estimate)
 export(util_gamma_stats_tbl)
+export(util_geometric_aic)
 export(util_geometric_param_estimate)
 export(util_geometric_stats_tbl)
 export(util_hypergeometric_param_estimate)

diff --git a/R/utils-aic-geometric.R b/R/utils-aic-geometric.R
@@ -0,0 +1,59 @@
+#' Calculate Akaike Information Criterion (AIC) for Geometric Distribution
+#'
+#' This function calculates the Akaike Information Criterion (AIC) for a geometric
+#' distribution fitted to the provided data.
+#'
+#' @family Utility
+#' @author Steven P. Sanderson II, MPH
+#'
+#' @description
+#' This function estimates the probability parameter of a geometric distribution
+#' from the provided data and then calculates the AIC value based on the fitted
+#' distribution.
+#'
+#' @param .x A numeric vector containing the data to be fitted to a geometric distribution.
+#'
+#' @details
+#' This function fits a geometric distribution to the provided data. It estimates
+#' the probability parameter of the geometric distribution from the data. Then,
+#' it calculates the AIC value based on the fitted distribution.
+#'
+#' Initial parameter estimates: The function uses the method of moments estimate
+#' as a starting point for the probability parameter of the geometric distribution.
+#'
+#' Optimization method: Since the parameter is directly calculated from the data,
+#' no optimization is needed.
+#'
+#' 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 <- rgeom(100, prob = 0.2)
+#' util_geometric_aic(x)
+#'
+#' @return
+#' The AIC value calculated based on the fitted geometric distribution to the provided data.
+#'
+#' @name util_geometric_aic
+NULL
+
+#' @export
+#' @rdname util_geometric_aic
+util_geometric_aic <- function(.x) {
+  # Tidyeval
+  x <- as.numeric(.x)
+
+  # Estimate probability parameter
+  prob <- 1 / mean(x)
+
+  # Calculate AIC
+  k_geometric <- 1 # Number of parameters for geometric distribution (prob)
+  logLik_geometric <- sum(dgeom(x, prob = prob, log = TRUE))
+  AIC_geometric <- 2 * k_geometric - 2 * logLik_geometric
+
+  # Return AIC
+  return(AIC_geometric)
+}
diff --git a/man/check_duplicate_rows.Rd b/man/check_duplicate_rows.Rd
diff --git a/man/convert_to_ts.Rd b/man/convert_to_ts.Rd
diff --git a/man/quantile_normalize.Rd b/man/quantile_normalize.Rd
diff --git a/man/tidy_mcmc_sampling.Rd b/man/tidy_mcmc_sampling.Rd
diff --git a/man/util_beta_aic.Rd b/man/util_beta_aic.Rd
diff --git a/man/util_binomial_aic.Rd b/man/util_binomial_aic.Rd
diff --git a/man/util_cauchy_aic.Rd b/man/util_cauchy_aic.Rd
diff --git a/man/util_chisq_aic.Rd b/man/util_chisq_aic.Rd
diff --git a/man/util_exponential_aic.Rd b/man/util_exponential_aic.Rd
diff --git a/man/util_gamma_aic.Rd b/man/util_gamma_aic.Rd
diff --git a/man/util_geometric_aic.Rd b/man/util_geometric_aic.Rd
diff --git a/man/util_logistic_aic.Rd b/man/util_logistic_aic.Rd
diff --git a/man/util_lognormal_aic.Rd b/man/util_lognormal_aic.Rd
diff --git a/man/util_normal_aic.Rd b/man/util_normal_aic.Rd
diff --git a/man/util_pareto_aic.Rd b/man/util_pareto_aic.Rd
diff --git a/man/util_uniform_aic.Rd b/man/util_uniform_aic.Rd
diff --git a/man/util_weibull_aic.Rd b/man/util_weibull_aic.Rd