Week_5_Lab_Abkurtzer

library(completejourney)
library(dplyr)
library(ggplot2)
library(lubridate)

transactions <- get_data("transactions")

Loading completejourney data sets from GitHub
Download complete. Learn more about these data sets at http://bit.ly/completejourney

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

# Manual axes
age_order <- c("19-24", "25-34", "35-44", "45-54", "55-64", "65+")
income_order <- c("Under 15K", "15-24K", "25-34K", "35-49K", "50-74K", "75-99K",
                  "100-124K", "125-149K", "150-174K", "175-199K", "200-249K", "250K+")

sales_demo <- sales_demo %>%
  mutate(age = factor(age, levels = age_order, ordered = TRUE),
         income = factor(income, levels = income_order, ordered = TRUE),
         year = year(transaction_timestamp))

# average sales for each income and age
avg_sales_per_year <- sales_demo %>%
  group_by(income, age, year) %>%
  summarise(sales_value = sum(sales_value, na.rm = TRUE), .groups = "drop")

# average sales for each pairing of income and age
avg_sales <- avg_sales_per_year %>%
  group_by(income, age) %>%
  summarise(sales_value = mean(sales_value, na.rm = TRUE), .groups = "drop")

# scale bubble size
max_avg_sales <- max(avg_sales$sales_value, na.rm = TRUE)
avg_sales <- avg_sales %>%
  mutate(bubble_size = (sales_value / max_avg_sales) * 500)


ggplot(avg_sales, aes(x = income, y = age, size = bubble_size, color = sales_value)) +
  geom_point(alpha = 0.8, stroke = 1) +
  scale_color_gradient(low = "lightgreen", high = "darkgreen") +
  labs(title = "Average Total Sales Per Year by Income and Age Range",
       x = "Income",
       y = "Age Range",
       color = "Total Sales") +
  guides(size = "none") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) 

library(tidyverse)


merged_data <- transactions %>%
  inner_join(demographics, by = "household_id") %>%
  inner_join(products, by = "product_id")

# separate kids vs no kids
merged_data <- merged_data %>%
  mutate(has_kids = ifelse(kids_count > 0, "Has Kids", "No Kids"))

# remove Gasoline purchases
filtered_data <- merged_data %>%
  filter(product_type != "GASOLINE-REG UNLEADED")

# group by products and kids
top_products <- filtered_data %>%
  group_by(has_kids, product_type) %>%
  summarise(total_purchases = sum(quantity, na.rm = TRUE)) %>%
  arrange(has_kids, desc(total_purchases)) %>%
  group_by(has_kids) %>%
  slice_head(n = 10)  

`summarise()` has grouped output by 'has_kids'. You can override using the `.groups` argument.

ggplot(top_products, aes(x = reorder(product_type, -total_purchases), y = total_purchases, fill = has_kids)) +
  geom_bar(stat = "identity", position = "dodge") +
  labs(title = "Top 10 Product Types Purchased by Households (Excluding Gasoline)",
       x = "Product Type",
       y = "Total Purchases") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

NA
NA

library(patchwork)

merged_data <- transactions %>%
  inner_join(demographics, by = "household_id") %>%
  filter(!is.na(retail_disc), retail_disc > 0, sales_value > 0)  

# Calculate discount percent
merged_data <- merged_data %>%
  mutate(discount_percent = (retail_disc / (sales_value + retail_disc)) * 100)  
        

# total sales per store
store_sales <- merged_data %>%
  group_by(store_id) %>%
  summarise(total_sales = sum(sales_value, na.rm = TRUE)) %>%
  arrange(desc(total_sales))  

# top 3 and bottom 3 stores
top_3_stores <- head(store_sales, 3)$store_id
bottom_3_stores <- tail(store_sales, 3)$store_id

# filter for top and bottom stores
filtered_data <- merged_data %>%
  filter(store_id %in% c(top_3_stores, bottom_3_stores)) %>%
  filter(discount_percent <= 50) %>%
  mutate(Total_Sales_Value_Performance = ifelse(store_id %in% top_3_stores, "Top 3 Stores", "Bottom 3 Stores"))


box_plot <- ggplot(filtered_data, aes(x = Total_Sales_Value_Performance, y = discount_percent, fill = Total_Sales_Value_Performance)) +
  geom_boxplot(outlier.color = "red", outlier.shape = 16) +
  scale_fill_manual(values = c("Bottom 3 Stores" = "orange", "Top 3 Stores" = "navyblue")) + 
  labs(title = "Distribution of Discount % of Stores with Highest and Lowest Total Sales Value",
       x = "Stores",
       y = "Discount Percentage") +
  theme_minimal()
box_plot