merge pr #410: standardize distribution naming

Author: simonpcouch

R/get_confidence_interval.R

@@ -64,15 +64,15 @@
 #'
 #' @examples
 #'
-#' boot_distr <- gss %>%
+#' boot_dist <- gss %>%
 #'   # We're interested in the number of hours worked per week
 #'   specify(response = hours) %>%
 #'   # Generate bootstrap samples
 #'   generate(reps = 1000, type = "bootstrap") %>%
 #'   # Calculate mean of each bootstrap sample
 #'   calculate(stat = "mean")
 #'
-#' boot_distr %>%
+#' boot_dist %>%
 #'   # Calculate the confidence interval around the point estimate
 #'   get_confidence_interval(
 #'     # At the 95% confidence level; percentile method
@@ -84,7 +84,7 @@
 #'   specify(response = hours) %>%
 #'   calculate(stat = "mean")
 #'
-#' boot_distr %>%
+#' boot_dist %>%
 #'   get_confidence_interval(
 #'     point_estimate = sample_mean,
 #'     # At the 95% confidence level

R/shade_p_value.R

@@ -55,11 +55,11 @@
 #'   
 #' # you can shade confidence intervals on top of
 #' # theoretical distributions, too!
-#' null_dist_theoretical <- gss %>%
+#' null_dist_theory <- gss %>%
 #'   specify(response = hours) %>%
 #'   assume(distribution = "t") 
 #'   
-#' null_dist_theoretical %>%
+#' null_dist_theory %>%
 #'   visualize() +
 #'   shade_p_value(obs_stat = point_estimate, direction = "two-sided")
 #' 

README.Rmd

@@ -86,7 +86,7 @@ F_hat <- gss %>%
 Then, generating the null distribution,
 
 ```{r, message = FALSE, warning = FALSE}
-null_distn <- gss %>%
+null_dist <- gss %>%
    specify(age ~ partyid) %>%
    hypothesize(null = "independence") %>%
    generate(reps = 1000, type = "permute") %>%
@@ -96,7 +96,7 @@ null_distn <- gss %>%
 Visualizing the observed statistic alongside the null distribution,
 
 ```{r viz, message = FALSE, warning = FALSE, eval = FALSE}
-visualize(null_distn) +
+visualize(null_dist) +
   shade_p_value(obs_stat = F_hat, direction = "greater")
 ```
 
@@ -107,7 +107,7 @@ knitr::include_graphics("https://raw.githubusercontent.com/tidymodels/infer/mast
 Calculating the p-value from the null distribution and observed statistic,
 
 ```{r, message = FALSE, warning = FALSE}
-null_distn %>%
+null_dist %>%
   get_p_value(obs_stat = F_hat, direction = "greater")
 ```
 

README.md

@@ -129,7 +129,7 @@ F_hat <- gss %>%
 Then, generating the null distribution,
 
 ``` r
-null_distn <- gss %>%
+null_dist <- gss %>%
    specify(age ~ partyid) %>%
    hypothesize(null = "independence") %>%
    generate(reps = 1000, type = "permute") %>%
@@ -139,7 +139,7 @@ null_distn <- gss %>%
 Visualizing the observed statistic alongside the null distribution,
 
 ``` r
-visualize(null_distn) +
+visualize(null_dist) +
   shade_p_value(obs_stat = F_hat, direction = "greater")
 ```
 
@@ -155,7 +155,7 @@ Calculating the p-value from the null distribution and observed
 statistic,
 
 ``` r
-null_distn %>%
+null_dist %>%
   get_p_value(obs_stat = F_hat, direction = "greater")
 ```
 

man/get_confidence_interval.Rd

@@ -255,7 +255,7 @@ test_that("theoretical CIs align with simulation-based (mean)", {
     generate(reps = 1e3, type = "bootstrap") %>% 
     calculate(stat = "mean")
 
-  null_dist_theor <- gss %>%
+  null_dist_theory <- gss %>%
     specify(response = hours) %>%
     hypothesize(null = "point", mu = 40) %>%
     assume(distribution = "t")
@@ -268,7 +268,7 @@ test_that("theoretical CIs align with simulation-based (mean)", {
       point_estimate = x_bar
     ),
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       .95, 
       type = "se", 
       point_estimate = x_bar
@@ -290,7 +290,7 @@ test_that("theoretical CIs align with simulation-based (prop)", {
     generate(reps = 1e3, type = "draw") %>%
     calculate(stat = "prop")
 
-  null_dist_theor <- gss %>%
+  null_dist_theory <- gss %>%
     specify(response = sex, success = "female") %>%
     assume(distribution = "z")
 
@@ -302,7 +302,7 @@ test_that("theoretical CIs align with simulation-based (prop)", {
       point_estimate = p_hat
     ),
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       .95, 
       type = "se", 
       point_estimate = p_hat
@@ -324,7 +324,7 @@ test_that("theoretical CIs align with simulation-based (diff in means)", {
     generate(reps = 3e3, type = "permute") %>% 
     calculate(stat = "diff in means", order = c("degree", "no degree"))
 
-  null_dist_theor <- gss %>%
+  null_dist_theory <- gss %>%
     specify(age ~ college) %>% 
     assume(distribution = "t")
 
@@ -336,7 +336,7 @@ test_that("theoretical CIs align with simulation-based (diff in means)", {
       point_estimate = diff_bar
     ),
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       .95, 
       type = "se", 
       point_estimate = diff_bar
@@ -358,7 +358,7 @@ test_that("theoretical CIs align with simulation-based (diff in props)", {
     generate(reps = 1e3, type = "permute") %>%
     calculate(stat = "diff in props", order = c("female", "male"))
 
-  null_dist_theor <- gss %>%
+  null_dist_theory <- gss %>%
     specify(college ~ sex, success = "no degree") %>%
     assume(distribution = "z")
 
@@ -370,7 +370,7 @@ test_that("theoretical CIs align with simulation-based (diff in props)", {
       point_estimate = diff_hat
     ),
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       .95, 
       type = "se", 
       point_estimate = diff_hat
@@ -384,19 +384,19 @@ test_that("theoretical CIs check arguments properly", {
     specify(response = hours) %>%
     calculate(stat = "mean")
 
-  null_dist_theor <- gss %>%
+  null_dist_theory <- gss %>%
     specify(age ~ college) %>% 
     assume(distribution = "t")
 
   # check that type is handled correctly
   expect_equal(
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       level = .95, 
       point_estimate = x_bar
     ),
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       level = .95,
       type = "se",
       point_estimate = x_bar
@@ -405,7 +405,7 @@ test_that("theoretical CIs check arguments properly", {
 
   expect_error(
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       level = .95,
       type = "percentile",
       point_estimate = x_bar
@@ -415,7 +415,7 @@ test_that("theoretical CIs check arguments properly", {
 
   expect_error(
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       level = .95,
       type = "boop",
       point_estimate = x_bar
@@ -426,7 +426,7 @@ test_that("theoretical CIs check arguments properly", {
   # check that point estimate hasn't been post-processed
   expect_error(
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       level = .95, 
       point_estimate = dplyr::pull(x_bar)
     ),
@@ -435,7 +435,7 @@ test_that("theoretical CIs check arguments properly", {
 
   expect_error(
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       level = .95, 
       point_estimate = x_bar$stat
     ),
@@ -450,7 +450,7 @@ test_that("theoretical CIs check arguments properly", {
 
   expect_error(
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       level = .95, 
       point_estimate = obs_t
     ),
@@ -468,7 +468,7 @@ test_that("theoretical CIs check arguments properly", {
 
   expect_error(
     get_confidence_interval(
-      null_dist_theor, 
+      null_dist_theory, 
       level = .95, 
       point_estimate = p_hat
     ),

man/shade_p_value.Rd

@@ -61,7 +61,7 @@ We can `generate` an approximation of the null distribution using randomization.
 
 ```{r generate-null-f, warning = FALSE, message = FALSE}
 # generate the null distribution using randomization
-null_distribution <- gss %>%
+null_dist <- gss %>%
   specify(age ~ partyid) %>%
   hypothesize(null = "independence") %>%
   generate(reps = 1000, type = "permute") %>%
@@ -74,7 +74,7 @@ To get a sense for what this distribution looks like, and where our observed sta
 
 ```{r visualize-f, warning = FALSE, message = FALSE}
 # visualize the null distribution and test statistic!
-null_distribution %>%
+null_dist %>%
   visualize() + 
   shade_p_value(observed_f_statistic,
                 direction = "greater")
@@ -84,11 +84,11 @@ We could also visualize the observed statistic against the theoretical null dist
 
 ```{r visualize-f-theor, warning = FALSE, message = FALSE}
 # visualize the theoretical null distribution and test statistic!
-null_dist_theoretical <- gss %>%
+null_dist_theory <- gss %>%
   specify(age ~ partyid) %>%
   assume(distribution = "F")
 
-visualize(null_dist_theoretical) +
+visualize(null_dist_theory) +
   shade_p_value(observed_f_statistic,
                 direction = "greater")
 ```
@@ -97,7 +97,7 @@ To visualize both the randomization-based and theoretical null distributions to
 
 ```{r visualize-indep-both, warning = FALSE, message = FALSE}
 # visualize both null distributions and the test statistic!
-null_distribution %>%
+null_dist %>%
   visualize(method = "both") + 
   shade_p_value(observed_f_statistic,
                 direction = "greater")
@@ -107,7 +107,7 @@ Either way, it looks like our observed test statistic would be quite unlikely if
 
 ```{r p-value-indep, warning = FALSE, message = FALSE}
 # calculate the p value from the observed statistic and null distribution
-p_value <- null_distribution %>%
+p_value <- null_dist %>%
   get_p_value(obs_stat = observed_f_statistic,
               direction = "greater")