BANA7025 Moduel 5 Lab Ang Zhang.Rmd

---
title: "BANA7025 Module 5 Lab"
author: "Ang Zhang"
output:
  html_document:
    df_print: paged
---



```{r message=FALSE, warning=FALSE}
packages <- c("ggplot2", "dplyr", "stringr", "completejourney", "lubridate")
pkg <- lapply(packages, library, character.only = T)
```

```{r message=FALSE, warning=FALSE}
transactions <- get_transactions()
promotions <- get_promotions()
df <- transactions %>%
  left_join(products, by = "product_id") %>%
  left_join(promotions, by = c("product_id", "store_id")) %>%
  mutate(on_display = ifelse(is.na(display_location), F, display_location !=0)) %>%
  mutate(on_mailer = ifelse(is.na(mailer_location), F, mailer_location !=0))

# try to investigate the effect of promotions on sales value as well as quantity. 
# fetch the data and give each a flag as whether related to in-store display promotion or mailer promotion.
df <- df %>%
  mutate(hour = hour(transaction_timestamp))
```

```{r}
df %>%
  count(hour) %>%
  ggplot(aes(x = hour, y = n, fill = n)) + 
  geom_bar(stat = "identity") +
  labs(title = "Customer Shopping Behavior Investigation", 
       subtitle = "number of shopping across time of the day\nshoppers peak around 5pm while few were seen before 10am. stores should staff accordingly.",
       x = "time (hour) of the day",
       y = "total number of transactions observed") + theme_minimal() +
  theme(
    plot.subtitle = element_text(hjust = 0, size = 10)
  )
```


```{r}
plot1 <- df %>%
  group_by(brand, department) %>%
  summarize(total_sales = sum(sales_value)) %>%
  treemap::treemap(
    index = c("brand", "department"),
    vSize = "total_sales",
    title = "Total sales generated by different deparments overlayed by brand type"
  )
  
```
```{r}
plot1_data <- df %>%
  group_by(on_display) %>%
  summarize(product_count = n_distinct(product_id)) %>%
  mutate(ypos = product_count - 0.2*product_count)
plot1 <- ggplot(plot1_data, aes(x = "", y = product_count, fill = on_display)) + 
  geom_bar(stat = "identity", width = 1, color = "white") + 
  coord_polar("y", start = 0) +
  theme_void() +
  theme(legend.position = "none") + 
  geom_text(aes(y = ypos, label = c("on display", "not on display"))) +
  ggtitle("Product constitution")
plot2_data <- df %>%
  group_by(on_display) %>%
  summarize(transaction_count = n()) %>%
  mutate(ypos = cumsum(transaction_count) - 0.5*transaction_count)
plot2 <- ggplot(plot2_data, aes(x = "", y = transaction_count, fill = on_display)) + 
  geom_bar(stat = "identity", width = 1, color = "white") + 
  coord_polar("y", start = 0) +
  theme_void() +
  theme(legend.position = "none") + 
  geom_text(aes(y = ypos, label = c("on display", "not on display"))) +
  ggtitle("Transaction contribution")
plot3_data <- df %>%
  group_by(on_display) %>%
  summarize(total_sales = sum(sales_value)) %>%
  mutate(ypos = cumsum(total_sales) - 0.5*total_sales)
plot3 <- ggplot(plot3_data, aes(x = "", y = total_sales, fill = on_display)) + 
  geom_bar(stat = "identity", width = 1, color = "white") + 
  coord_polar("y", start = 0) +
  theme_void() +
  theme(legend.position = "none") + 
  geom_text(aes(y = ypos, label = c("on display", "not on display"))) +
  ggtitle("Sales contribution")
gridExtra::grid.arrange(
  grid::textGrob("On display product contribution in sales & transaction", gp = grid::gpar(fontsize = 15, fontface = "bold")),
  grid::textGrob("On display products account for less than 1/4 but contributed nearly half of total sales & transaction", gp = grid::gpar(fontsize = 10)),
  gridExtra::arrangeGrob(plot1, plot2, plot3, nrow = 1),
  heights = c(0.1, 0.05, 0.60)
  )


```