## ----hypothesis-testing-load-packages, message=FALSE-------------------------- library(tidyverse) library(moderndive) library(infer) library(nycflights23) library(ggplot2movies) ## ----hypothesis-testing-alt2-------------------------------------------------- almonds_sample_100 ## ----hypothesis-testing-mean-and-sd------------------------------------------- almonds_sample_100 |> summarize(sample_mean = mean(weight), sample_sd = sd(weight)) ## ----hypothesis-testing-mean-sd, eval=FALSE----------------------------------- # almonds_sample_100 |> # summarize(x_bar = mean(weight), # s = sd(weight), # n = length(weight), # t = (x_bar - 3.6)/(s/sqrt(n))) ## ----hypothesis-testing-demo-code, eval = FALSE------------------------------- # 2 * pt(q = -2.26, df = 100 - 1) ## ----hypothesis-testing-null-dist--------------------------------------------- null_dist <- almonds_sample_100 |> specify(response = weight) |> hypothesize(null = "point", mu = 3.6) |> generate(reps = 1000, type = "bootstrap") |> calculate(stat = "mean") ## ----hypothesis-testing-create-x_bar_almonds, echo=TRUE----------------------- x_bar_almonds <- almonds_sample_100 |> summarize(sample_mean = mean(weight)) |> select(sample_mean) null_dist |> get_p_value(obs_stat = x_bar_almonds, direction = "two-sided") ## ----hypothesis-testing-create-p_val_almonds, echo=FALSE---------------------- p_val_almonds <- null_dist |> get_p_value(obs_stat = x_bar_almonds, direction = "two-sided") ## ----hypothesis-testing-mean-sd-v2-dup1--------------------------------------- almonds_sample_100 |> summarize(lower_bound = mean(weight) - 1.98*sd(weight)/sqrt(length(weight)), upper_bound = mean(weight) + 1.98*sd(weight)/sqrt(length(weight))) ## ----hypothesis-testing-bootstrap, eval=FALSE--------------------------------- # bootstrap_means <- almonds_sample_100 |> # specify(response = weight) |> # generate(reps = 1000, type = "bootstrap") |> # calculate(stat = "mean") ## ----hypothesis-testing-conf-interval----------------------------------------- bootstrap_means |> get_confidence_interval(level = 0.95, type = "percentile") ## ----hypothesis-testing-v17, echo=FALSE--------------------------------------- set.seed(2) ## ----hypothesis-testing-create-spotify_metal_deep, eval=FALSE----------------- # spotify_metal_deephouse <- spotify_by_genre |> # filter(track_genre %in% c("metal", "deep-house")) |> # select(track_genre, artists, track_name, popularity, popular_or_not) # spotify_metal_deephouse |> # group_by(track_genre, popular_or_not) |> # sample_n(size = 3) ## ----twelve-spotify, echo=FALSE----------------------------------------------- spotify_metal_deephouse <- spotify_by_genre |> filter(track_genre %in% c("metal", "deep-house")) |> select(track_id, track_genre, artists, track_name, popularity, popular_or_not) sampled_spotify_metal_deephouse <- spotify_metal_deephouse |> group_by(track_genre, popular_or_not) |> sample_n(size = 3) |> arrange(track_id) |> ungroup() |> select(-track_id) sampled_spotify_metal_deephouse |> kbl( caption = "Sample of twelve songs from the Spotify data frame.", booktabs = TRUE, linesep = "" ) |> kable_styling( font_size = ifelse(is_latex_output(), 8, 16), latex_options = c("HOLD_position") ) |> column_spec(3, width = "1.5in") # Adjust the column number/width as needed ## ----hypothesis-testing-bar, eval=FALSE--------------------------------------- # ggplot(spotify_metal_deephouse, aes(x = track_genre, fill = popular_or_not)) + # geom_bar() + # labs(x = "Genre of track") ## ----hypothesis-testing-grouped-summary--------------------------------------- spotify_metal_deephouse |> group_by(track_genre, popular_or_not) |> tally() # Same as summarize(n = n()) ## ----hypothesis-testing-select-vars, eval=FALSE------------------------------- # spotify_52_original |> # select(-track_id) |> # head(10) ## ----spotify-52-sample, echo=FALSE-------------------------------------------- spotify_52_original |> select(-track_id) |> head(10) |> kbl(caption = "Representative sample of metal and deep-house songs", booktabs = TRUE, linesep = "" ) |> kable_styling( font_size = ifelse(is_latex_output(), 8, 16), latex_options = c("HOLD_position") ) |> column_spec(3, width = "1.5in") # Adjust the column number and width as needed ## ----hypothesis-testing-select-vars-alt, eval=FALSE--------------------------- # spotify_52_shuffled |> # select(-track_id) |> # head(10) ## ----spotify-shuffled-52-sample, echo=FALSE----------------------------------- spotify_52_shuffled |> select(-track_id) |> head(10) |> kbl(caption = "(ref:spotify-shuffled-52)", booktabs = TRUE, linesep = "" ) |> kable_styling( font_size = ifelse(is_latex_output(), 8, 16), latex_options = c("HOLD_position") ) |> column_spec(3, width = "1.5in") ## ----hypothesis-testing-bar-filled, eval=FALSE-------------------------------- # ggplot(spotify_52_shuffled, aes(x = track_genre, fill = popular_or_not)) + # geom_bar() + # labs(x = "Genre of track") ## ----hypothesis-testing-v26--------------------------------------------------- spotify_52_shuffled |> group_by(track_genre, popular_or_not) |> tally() ## ----hypothesis-testing-specify----------------------------------------------- spotify_metal_deephouse |> specify(formula = popular_or_not ~ track_genre, success = "popular") ## ----hypothesis-testing-null-dist-sized--------------------------------------- spotify_metal_deephouse |> specify(formula = popular_or_not ~ track_genre, success = "popular") |> hypothesize(null = "independence") ## ----hypothesis-testing-null-dist-alt2, eval=FALSE---------------------------- # spotify_generate <- spotify_metal_deephouse |> # specify(formula = popular_or_not ~ track_genre, success = "popular") |> # hypothesize(null = "independence") |> # generate(reps = 1000, type = "permute") # nrow(spotify_generate) ## ----hypothesis-testing-null-dist-v5, eval=FALSE------------------------------ # null_distribution <- spotify_metal_deephouse |> # specify(formula = popular_or_not ~ track_genre, success = "popular") |> # hypothesize(null = "independence") |> # generate(reps = 1000, type = "permute") |> # calculate(stat = "diff in props", order = c("metal", "deep-house")) # null_distribution ## ----hypothesis-testing-specify-alt------------------------------------------- obs_diff_prop <- spotify_metal_deephouse |> specify(formula = popular_or_not ~ track_genre, success = "popular") |> calculate(stat = "diff in props", order = c("metal", "deep-house")) obs_diff_prop ## ----hypothesis-testing-v37--------------------------------------------------- spotify_metal_deephouse |> observe(formula = popular_or_not ~ track_genre, success = "popular", stat = "diff in props", order = c("metal", "deep-house")) ## ----null-distribution-infer, fig.show="hold", fig.cap="Null distribution.", fig.height=ifelse(knitr::is_latex_output(), 1.8, 4)---- visualize(null_distribution, bins = 25) ## ----null-distribution-infer-2, fig.cap="Shaded histogram to show $p$-value."---- visualize(null_distribution, bins = 25) + shade_p_value(obs_stat = obs_diff_prop, direction = "right") ## ----hypothesis-testing-v38--------------------------------------------------- null_distribution |> get_p_value(obs_stat = obs_diff_prop, direction = "right") ## ----hypothesis-testing-null-dist-v6, eval=FALSE------------------------------ # null_distribution <- spotify_metal_deephouse |> # specify(formula = popular_or_not ~ track_genre, success = "popular") |> # hypothesize(null = "independence") |> # generate(reps = 1000, type = "permute") |> # calculate(stat = "diff in props", order = c("metal", "deep-house")) ## ----hypothesis-testing-null-dist-v7, eval=FALSE------------------------------ # bootstrap_distribution <- spotify_metal_deephouse |> # specify(formula = popular_or_not ~ track_genre, success = "popular") |> # # Change 1 - Remove hypothesize(): # # hypothesize(null = "independence") |> # # Change 2 - Switch type from "permute" to "bootstrap": # generate(reps = 1000, type = "bootstrap") |> # calculate(stat = "diff in props", order = c("metal", "deep-house")) ## ----hypothesis-testing-assign-percentile_ci---------------------------------- percentile_ci <- bootstrap_distribution |> get_confidence_interval(level = 0.90, type = "percentile") percentile_ci ## ----hypothesis-testing-viz-dist, eval=FALSE---------------------------------- # visualize(bootstrap_distribution) + # shade_confidence_interval(endpoints = percentile_ci) ## ----hypothesis-testing-conf-interval-alt2, eval=FALSE------------------------ # se_ci <- bootstrap_distribution |> # get_confidence_interval(level = 0.95, type = "se", # point_estimate = obs_diff_prop) # se_ci ## ----hypothesis-testing-viz-dist-alt, eval=FALSE------------------------------ # visualize(bootstrap_distribution) + # shade_confidence_interval(endpoints = se_ci) ## ----hypothesis-testing-load-packages-sized, eval=FALSE----------------------- # library(moderndive) # library(infer) # null_distribution_mean <- spotify_metal_deephouse |> # specify(formula = popular_or_not ~ track_genre, success = "popular") |> # hypothesize(null = "independence") |> # generate(reps = 1000, type = "permute") |> # calculate(stat = "diff in means", order = c("metal", "deep-house")) ## ----hypothesis-testing-v50--------------------------------------------------- movies ## ----hypothesis-testing-v52--------------------------------------------------- movies_sample ## ----action-romance-boxplot, fig.cap="Boxplot of IMDb rating vs. genre.", fig.height=ifelse(knitr::is_latex_output(), 4, 4)---- ggplot(data = movies_sample, aes(x = genre, y = rating)) + geom_boxplot() + labs(y = "IMDb rating") ## ----hypothesis-testing-mean-sd-v2-dup2--------------------------------------- movies_sample |> group_by(genre) |> summarize(n = n(), mean_rating = mean(rating), std_dev = sd(rating)) ## ----hypothesis-testing-specify-alt2------------------------------------------ movies_sample |> specify(formula = rating ~ genre) ## ----hypothesis-testing-null-dist-v9------------------------------------------ movies_sample |> specify(formula = rating ~ genre) |> hypothesize(null = "independence") ## ----hypothesis-testing-view, eval=FALSE-------------------------------------- # movies_sample |> # specify(formula = rating ~ genre) |> # hypothesize(null = "independence") |> # generate(reps = 1000, type = "permute") |> # View() ## ----hypothesis-testing-null-dist-v11, eval=FALSE----------------------------- # null_distribution_movies <- movies_sample |> # specify(formula = rating ~ genre) |> # hypothesize(null = "independence") |> # generate(reps = 1000, type = "permute") |> # calculate(stat = "diff in means", order = c("Action", "Romance")) # null_distribution_movies ## ----hypothesis-testing-specify-v4-------------------------------------------- obs_diff_means <- movies_sample |> specify(formula = rating ~ genre) |> calculate(stat = "diff in means", order = c("Action", "Romance")) obs_diff_means ## ----hypothesis-testing-viz-pvalue, eval=FALSE-------------------------------- # visualize(null_distribution_movies, bins = 10) + # shade_p_value(obs_stat = obs_diff_means, direction = "both") ## ----hypothesis-testing-v63--------------------------------------------------- null_distribution_movies |> get_p_value(obs_stat = obs_diff_means, direction = "both") ## ----hypothesis-testing-create-p_value_movies, echo=FALSE--------------------- p_value_movies <- null_distribution_movies |> get_p_value(obs_stat = obs_diff_means, direction = "both") |> mutate(p_value = round(p_value, 3)) ## ----hypothesis-testing-mean-sd-v2-dup3--------------------------------------- movies_sample |> group_by(genre) |> summarize(n = n(), mean_rating = mean(rating), std_dev = sd(rating)) ## ----hypothesis-testing-v67--------------------------------------------------- movies_sample |> t_test(formula = rating ~ genre, order = c("Action", "Romance"), alternative = "two-sided") ## ----hypothesis-testing-filter-alaska----------------------------------------- flights_sample <- flights |> filter(carrier %in% c("HA", "AS")) ## ----ha-as-flights-boxplot, fig.cap="Air time for Hawaiian and Alaska Airlines flights departing NYC in 2023.", fig.height=ifelse(knitr::is_latex_output(), 2.8, 4)---- ggplot(data = flights_sample, mapping = aes(x = carrier, y = air_time)) + geom_boxplot() + labs(x = "Carrier", y = "Air Time") ## ----hypothesis-testing-summary-by-carrier------------------------------------ flights_sample |> group_by(carrier, dest) |> summarize(n = n(), mean_time = mean(air_time, na.rm = TRUE), .groups = "keep")