Module 5 - Visualization Lab

Which Product Categories are the Most Popular

## Which Product Categories are the Most Popular?

joined_data2 <- transactions %>%
  inner_join(products, by = "product_id")

popular_categories <- joined_data2 %>%
  group_by(product_category) %>%
  summarize(transaction_count = n()) %>%
  arrange(desc(transaction_count))

top_10_categories <- popular_categories %>%
  slice_head(n = 10)

ggplot(top_10_categories, aes(x = reorder(product_category, transaction_count), y = transaction_count)) +
  geom_bar(stat = "identity", fill = "skyblue") +
  coord_flip() +
  labs(
    title = "Most Popular Product Categories",
    x = "Product Category",
    y = "Number of Transactions"
  ) +
  theme_classic()

What is the Average Candy Sales and Is There a Trend by Day of the Week?

## What is the Average Candy Sales and Is There a Trend by Day of the Week?

joined_data1 <- products %>%
  inner_join(transactions)

## Joining with `by = join_by(product_id)`

candy <- joined_data1 %>%
  filter(str_detect(product_category, regex("CANDY - CHECKLANE", ignore_case = TRUE))) %>%
  filter(!product_type %in% c("SEASONAL CANDY BAGS-CHOCOLATE", "NOVELTY CANDY", "MISC CHECKLANE CANDY", 
                                "CANDY & BREATH MINTS (PKGD) (N"))
candy_sales <- candy %>%
  inner_join(transactions) %>%
  mutate(day = lubridate::wday(transaction_timestamp, label = TRUE)) %>%
  group_by(day, product_type) %>%
  summarize(total_sales = sum(sales_value, na.rm = TRUE), .groups = 'drop')

## Joining with `by = join_by(product_id, household_id, store_id, basket_id,
## quantity, sales_value, retail_disc, coupon_disc, coupon_match_disc, week,
## transaction_timestamp)`

ggplot(candy_sales, aes(x = day, y = total_sales, fill = product_type)) +
  geom_bar(stat = "identity", position = "dodge") +
  scale_y_continuous(labels = scales::dollar) +
  labs(title = "Total Candy Sales by Day of the Week and Product Category", 
       x = "Day of the Week", 
       y = "Total Sales Value") +
  theme_bw() +
  scale_fill_manual(values = my_colors)

How Does Household Composition Affect Basket Size and Frequency of Purchases?

## How Does Household Composition Affect Basket Size and Frequency of Purchases?

joined_data <- transactions %>%
  inner_join(demographics, by = "household_id") 

summary_stats <- joined_data %>%
  group_by(household_id) %>%
  summarize(total_sales = sum(sales_value, na.rm = TRUE))

frequency_of_purchases <- joined_data %>%
  group_by(household_id) %>%
  summarize(purchase_frequency = n_distinct(basket_id))

basket_size <- summary_stats %>%
  left_join(frequency_of_purchases, by = "household_id") %>%
  left_join(demographics %>% select(household_id, age), by = "household_id")

ggplot(basket_size, aes(x = purchase_frequency, y = total_sales, fill = age)) +
  geom_boxplot(alpha = .5) +
  scale_fill_manual(values = my_colors1) +
  facet_grid(~ age) + 
  labs(
    title = "Total Sales vs. Frequency of Purchases",
    x = "Frequency of Purchases",
    y = "Total Sales") + 
  theme_linedraw()