Merge pull request #456 from spsanderson/development

prepping for CRAN

R/00_global_variables.R

@@ -7,6 +7,6 @@ globalVariables(
     "name", "prob", "rate", "sd_log", "shape1", "shape2", "size", "total", "total_deviance",
     "abs_aic", "aic_value", "data", "dist_with_params", "ks", "lm", "lm_model", "mu",
     "stan_dev", "tidy_ks", "prop", "dens_tbl", "cih", "cil", "mstat", "stat", ".",
-    "variable",".N"
+    "variable",".N",".cum_stat",".return_tibble",".sample","dof","ncp"
   )
 )

R/est-param-chisq.R

@@ -127,15 +127,15 @@ util_chisquare_param_estimate <- function(.x, .auto_gen_empirical = TRUE) {
   neg_log_likelihood <- function(params) {
     df <- params[1]
     ncp <- params[2]
-    sum_densities <- sum(dchisq(x_term, df = df, ncp = ncp, log = TRUE))
+    sum_densities <- sum(stats::dchisq(x_term, df = df, ncp = ncp, log = TRUE))
     return(-sum_densities)
   }
 
   # Initial guess for parameters
   initial_params <- c(trunc(var(x_term)/2), trunc(mean(x_term)))
 
   # Optimize parameters using optim() function
-  opt_result <- optim(par = initial_params, fn = neg_log_likelihood)
+  opt_result <- stats::optim(par = initial_params, fn = neg_log_likelihood)
 
   # Extract estimated parameters
   doff <- opt_result$par[1]

R/utils-aic-beta.R

@@ -60,11 +60,11 @@ util_beta_aic <- function(.x) {
     shape2 <- par[2]
     ncp <- par[3]
     n <- length(data)
-    -sum(dbeta(data, shape1, shape2, ncp, log = TRUE))
+    -sum(stats::dbeta(data, shape1, shape2, ncp, log = TRUE))
   }
 
   # Fit beta distribution using optim
-  fit_beta <- optim(
+  fit_beta <- stats::optim(
     c(trunc(pe$shape1), trunc(pe$shape2), 0),
     neg_log_lik_beta,
     data = x

R/utils-aic-cauchy.R

@@ -57,15 +57,15 @@ util_cauchy_aic <- function(.x) {
     location <- par[1]
     scale <- par[2]
     n <- length(data)
-    -sum(dcauchy(data, location = location, scale = scale, log = TRUE))
+    -sum(stats::dcauchy(data, location = location, scale = scale, log = TRUE))
   }
 
   # Get initial parameter estimates (you might need to adjust this depending on your data)
   # Here we use method of moments estimates as a starting point
   pe <- TidyDensity::util_cauchy_param_estimate(x)$parameter_tbl
 
   # Fit Cauchy distribution using optim
-  fit_cauchy <- optim(
+  fit_cauchy <- stats::optim(
     c(pe$location, pe$scale),
     neg_log_lik_cauchy,
     data = x

R/utils-aic-chisq.R

@@ -28,7 +28,7 @@ util_chisq_aic <- function(.x) {
   x <- as.numeric(.x)
 
   # Get parameters
-  pe <- TidyDensity::util_chisquare_param_estimate(x)$parameter_tbl |> head(1)
+  pe <- TidyDensity::util_chisquare_param_estimate(x)$parameter_tbl |> utils::head(1)
 
   # Negative log-likelihood function for chi-square distribution
   neg_log_lik_chisq <- function(par, data) {
@@ -39,7 +39,7 @@ util_chisq_aic <- function(.x) {
   }
 
   # Fit chi-square distribution to sample data (rchisq)
-  fit_chisq <- optim(
+  fit_chisq <- stats::optim(
     c(pe$degrees_of_freedom, pe$ncp),
     neg_log_lik_chisq,
     data = x

R/utils-aic-exponential.R

@@ -45,14 +45,14 @@ util_exponential_aic <- function(.x) {
   neg_log_lik_exponential <- function(par, data) {
     rate <- par[1]
     n <- length(data)
-    -sum(dexp(data, rate = rate, log = TRUE))
+    -sum(stats::dexp(data, rate = rate, log = TRUE))
   }
 
   # Get initial parameter estimate: reciprocal of the mean of the data
   pe <- TidyDensity::util_exponential_param_estimate(x)$parameter_tbl
 
   # Fit exponential distribution using optim
-  fit_exponential <- optim(
+  fit_exponential <- stats::optim(
     pe$rate,
     neg_log_lik_exponential,
     data = x,

R/utils-aic-gamma.R

@@ -54,15 +54,15 @@ util_gamma_aic <- function(.x) {
     shape <- par[1]
     scale <- par[2]
     n <- length(data)
-    -sum(dgamma(data, shape = shape, scale = scale, log = TRUE))
+    -sum(stats::dgamma(data, shape = shape, scale = scale, log = TRUE))
   }
 
   # Get initial parameter estimates: method of moments
   pe <- TidyDensity::util_gamma_param_estimate(x)$parameter_tbl |>
     subset(method == "EnvStats_MMUE")
 
   # Fit gamma distribution using optim
-  fit_gamma <- optim(
+  fit_gamma <- stats::optim(
     c(pe$shape, pe$scale),
     neg_log_lik_gamma,
     data = x

R/utils-aic-geometric.R

@@ -51,7 +51,7 @@ util_geometric_aic <- function(.x) {
 
   # Calculate AIC
   k_geometric <- 1 # Number of parameters for geometric distribution (prob)
-  logLik_geometric <- sum(dgeom(x, prob = prob, log = TRUE))
+  logLik_geometric <- sum(stats::dgeom(x, prob = prob, log = TRUE))
   AIC_geometric <- 2 * k_geometric - 2 * logLik_geometric
 
   # Return AIC

R/utils-aic-hypergeometric.R

@@ -55,7 +55,7 @@ util_hypergeometric_aic <- function(.x) {
 
   # Calculate AIC
   k_hypergeometric <- 3 # Number of parameters for hypergeometric distribution (m, n, and k)
-  logLik_hypergeometric <- sum(dhyper(x, m = m, n = n, k = k, log = TRUE))
+  logLik_hypergeometric <- sum(stats::dhyper(x, m = m, n = n, k = k, log = TRUE))
   AIC_hypergeometric <- 2 * k_hypergeometric - 2 * logLik_hypergeometric
 
   # Return AIC

R/utils-aic-logistic.R

@@ -53,15 +53,15 @@ util_logistic_aic <- function(.x) {
     location <- par[1]
     scale <- par[2]
     n <- length(data)
-    -sum(dlogis(data, location = location, scale = scale, log = TRUE))
+    -sum(stats::dlogis(data, location = location, scale = scale, log = TRUE))
   }
 
   # Get initial parameter estimates: method of moments
   pe <- TidyDensity::util_logistic_param_estimate(x)$parameter_tbl |>
     subset(method == "EnvStats_MLE")
 
   # Fit logistic distribution using optim
-  fit_logistic <- optim(
+  fit_logistic <- stats::optim(
     c(pe$location, pe$scale),
     neg_log_lik_logistic,
     data = x

R/utils-aic-lognormal.R

@@ -52,7 +52,7 @@ util_lognormal_aic <- function(.x) {
     meanlog <- par[1]
     sdlog <- par[2]
     n <- length(data)
-    -sum(dlnorm(data, meanlog = meanlog, sdlog = sdlog, log = TRUE))
+    -sum(stats::dlnorm(data, meanlog = meanlog, sdlog = sdlog, log = TRUE))
   }
 
   # Get initial parameter estimates: method of moments
@@ -62,7 +62,7 @@ util_lognormal_aic <- function(.x) {
   sdlog_est <- sqrt(m2 - m1^2)
 
   # Fit log-normal distribution using optim
-  fit_lognormal <- optim(
+  fit_lognormal <- stats::optim(
     c(meanlog_est, sdlog_est),
     neg_log_lik_lognormal,
     data = x

R/utils-aic-normal.R

@@ -29,7 +29,7 @@ util_normal_aic <- function(.x) {
   x <- as.numeric(.x)
 
   # Get parameters
-  pe <- TidyDensity::util_normal_param_estimate(x)$parameter_tbl |> head(1)
+  pe <- TidyDensity::util_normal_param_estimate(x)$parameter_tbl |> utils::head(1)
 
   # Negative log-likelihood function for normal distribution
   neg_log_lik_norm <- function(par, data) {
@@ -40,7 +40,7 @@ util_normal_aic <- function(.x) {
   }
 
   # Fit normal distribution to population data (rnorm)
-  fit_norm <- optim(
+  fit_norm <- stats::optim(
     c(pe$mu, pe$stan_dev),
     neg_log_lik_norm,
     data = x

R/utils-aic-pareto.R

@@ -60,7 +60,7 @@ util_pareto_aic <- function(.x) {
     subset(method == "MLE")
 
   # Fit Pareto distribution using optim
-  fit_pareto <- optim(
+  fit_pareto <- stats::optim(
     c(pe$shape, pe$scale),
     neg_log_lik_pareto,
     data = x

R/utils-aic-poisson.R

@@ -50,7 +50,7 @@ util_poisson_aic <- function(.x) {
 
   # Calculate AIC
   k_poisson <- 1 # Number of parameters for Poisson distribution (lambda)
-  logLik_poisson <- sum(dpois(x, lambda = lambda, log = TRUE))
+  logLik_poisson <- sum(stats::dpois(x, lambda = lambda, log = TRUE))
   AIC_poisson <- 2 * k_poisson - 2 * logLik_poisson
 
   # Return AIC

R/utils-aic-weibull.R

@@ -53,14 +53,14 @@ util_weibull_aic <- function(.x) {
     shape <- par[1]
     scale <- par[2]
     n <- length(data)
-    -sum(dweibull(data, shape = shape, scale = scale, log = TRUE))
+    -sum(stats::dweibull(data, shape = shape, scale = scale, log = TRUE))
   }
 
   # Get initial parameter estimates: method of moments
   pe <- TidyDensity::util_weibull_param_estimate(x)$parameter_tbl
 
   # Fit Weibull distribution using optim
-  fit_weibull <- optim(
+  fit_weibull <- stats::optim(
     c(pe$shape, pe$scale),
     neg_log_lik_weibull,
     data = x

README.md

@@ -50,18 +50,18 @@ library(ggplot2)
 
 tidy_normal()
 #> # A tibble: 50 × 7
-#>    sim_number     x      y    dx       dy      p      q
-#>    <fct>      <int>  <dbl> <dbl>    <dbl>  <dbl>  <dbl>
-#>  1 1              1  0.261 -3.68 0.000208 0.603   0.261
-#>  2 1              2  0.491 -3.53 0.000571 0.688   0.491
-#>  3 1              3  0.220 -3.37 0.00139  0.587   0.220
-#>  4 1              4 -0.801 -3.21 0.00303  0.212  -0.801
-#>  5 1              5  0.732 -3.06 0.00591  0.768   0.732
-#>  6 1              6 -1.87  -2.90 0.0104   0.0308 -1.87 
-#>  7 1              7 -0.898 -2.74 0.0168   0.185  -0.898
-#>  8 1              8  0.157 -2.59 0.0252   0.562   0.157
-#>  9 1              9 -1.06  -2.43 0.0359   0.145  -1.06 
-#> 10 1             10  0.411 -2.27 0.0495   0.660   0.411
+#>    sim_number     x      y    dx       dy     p      q
+#>    <fct>      <int>  <dbl> <dbl>    <dbl> <dbl>  <dbl>
+#>  1 1              1  0.227 -2.97 0.000238 0.590  0.227
+#>  2 1              2  1.12  -2.84 0.000640 0.869  1.12 
+#>  3 1              3  1.26  -2.71 0.00153  0.897  1.26 
+#>  4 1              4  0.204 -2.58 0.00326  0.581  0.204
+#>  5 1              5  1.04  -2.44 0.00620  0.852  1.04 
+#>  6 1              6 -0.180 -2.31 0.0106   0.429 -0.180
+#>  7 1              7  0.299 -2.18 0.0167   0.618  0.299
+#>  8 1              8  1.73  -2.04 0.0243   0.959  1.73 
+#>  9 1              9 -0.770 -1.91 0.0338   0.221 -0.770
+#> 10 1             10  0.385 -1.78 0.0463   0.650  0.385
 #> # ℹ 40 more rows
 ```