imdb_post_code.r

library(tidyverse)
library(here)
library(broom)
library(corrr)
library(forcats)
library(stringr)
library(lubridate)
library(gridExtra)

df <- read_csv("movie_metadata.csv")
df %>% glimpse()

imdb_score_summary <- df %>% 
  summarise_each(funs(min, max, mean, median, sd), imdb_score) %>% 
  gather(stat, value) 
  
df %>% 
  ggplot(aes(imdb_score)) + 
  geom_histogram(bins = 50, colour = "black", fill = "red", alpha = 0.4) +
  theme_minimal() +
  annotation_custom(tableGrob(imdb_score_summary, rows = NULL), 
                    xmin = 2.5, 
                    xmax = 4, 
                    ymin = 300, ymax = 400) +
  labs(title = "IMDB score distribution")

df %>% 
  group_by(title_year) %>% 
  summarise(mean_rating = mean(imdb_score), 
            upper_rating = quantile(imdb_score, 0.975), 
            lower_rating = quantile(imdb_score, 0.0275)) %>% 
  ggplot(aes(title_year, mean_rating)) +
  geom_line(colour = "dodger blue") + 
  geom_point(alpha = 0.5) + 
  geom_smooth(method = "lm", colour = "red", alpha = 0.6, se = FALSE) +
  geom_errorbar(aes(ymin = upper_rating, ymax = lower_rating)) + 
  geom_rug() +
  theme_minimal() +
  labs(title = "Average IMDB movie rating per year", 
       x = "Year", 
       y = "IMDB rating")

df %>% map_df(~sum(is.na(.))) %>% glimpse()

replace_na_median <- function(x){
  x[is.na(x)] <- median(x, na.rm = TRUE)
  x
}

num_df <- df %>% 
  map_if(is_numeric, replace_na_median) %>% 
  as_data_frame() %>% 
  select_if(is_numeric)

num_df <- num_df %>% 
  mutate(years_since_release = 2017 - title_year) %>% 
  select(-title_year)

correlations <- corrr::correlate(num_df) %>%
    gather(variable, correlation, 2:16) %>%
    select(rowname, variable, correlation) %>% 
    mutate(high_correlation = ifelse(abs(correlation) > 0.50, "high", "not so high"))
  
correlations %>% 
  filter(abs(correlation) > 0.5) %>% 
  arrange(rowname)

correlations %>% 
  ggplot(aes(reorder(rowname, correlation), reorder(variable, correlation), fill = correlation)) + 
  geom_tile(alpha = 0.6, colour = "black") +
  geom_text(aes(label = round(correlation, 2), colour = high_correlation)) +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 70, hjust = 1)) +
  scale_fill_gradient()

num_df <- num_df %>% 
  select(-actor_1_facebook_likes, 
         -actor_2_facebook_likes, 
         -actor_3_facebook_likes, 
         -num_voted_users, 
         -movie_facebook_likes)

fit <- num_df %>% 
  map(scale) %>%
  as.data.frame() %>% 
  bootstrap(10) %>% 
  do(tidy(lm(imdb_score ~., data = .), conf.int = TRUE)) %>% 
  mutate(sig_0005 = p.value < 0.005)

fit %>% 
  filter(term != "(Intercept)") %>% 
  group_by(term) %>% 
  ggplot(aes(reorder(term, estimate), estimate, colour = sig_0005)) + 
  geom_point(alpha = 0.4) +
  coord_flip() +
  theme_minimal() +
  geom_hline(aes(yintercept = 0), linetype = "dashed", colour = "blue") +
  labs(title = "Linear regression estimates", 
       subtitle = "Dependent variable: IMDB score", 
       x = "Independent variable", 
       y = "estimate")

fit %>% 
  filter(term != "(Intercept)") %>% 
  group_by(term) %>% 
  summarise(estimate = mean(estimate), 
            conf.low = mean(conf.low), 
            conf.high = mean(conf.high), 
            sig_0005 = ifelse(sum(sig_0005) > 6, "significant (< 0.005)", "not significant")) %>% 
  ggplot(aes(y = reorder(term, estimate), x = estimate, colour = sig_0005)) +
  geom_point() + 
  geom_errorbarh(aes(xmax = conf.high, xmin = conf.low)) +
  theme_minimal() +
  geom_vline(aes(xintercept = 0), linetype = "dashed", colour = "blue") +
  labs(title = "Linear regression estimates", 
       subtitle = "Dependent variable: IMDB score", 
       x = "Independent variable", 
       y = "estimate")