Data Exploration

Exercises ~ Week 3

Logo

1 Exercise 1

The following table shows sample information for three students. Each observation represents a single student and includes details such as their unique student ID, name, age, total credits completed, major field of study, and year level.

This dataset demonstrates a mixture of variable types:

  • Nominal: StudentID, Name, Major
  • Numeric: Age (continuous), CreditsCompleted (discrete)
  • Ordinal: YearLevel (Freshman → Senior)
StudentID Name Age CreditsCompleted Major YearLevel
S001 Alice 20 45 Data Sains Sophomore
S002 Budi 21 60 Mathematics Junior
S003 Citra 19 30 Statistics Freshman 
# 1. Create vectors for each variable
StudentID <- c("S001", "S002", "S003")       # Nominal / ID
Name <- c("Alice", "Budi", "Citra")          # Nominal / Name
Age <- c(20, 21, 19)                         # Numeric / Continuous
CreditsCompleted <- c(45, 60, 30)            # Numeric / Discrete

# Nominal
Major <- c("Data Sains", "Mathematics", "Statistics")  

# Ordinal
YearLevel <- factor(c("Sophomore", "Junior", "Freshman"),
                    levels = c("Freshman","Sophomore","Junior","Senior"),
                    ordered = TRUE)          

# 2. Combine all vectors into a data frame
students <- data.frame(
  StudentID, Name, Age, CreditsCompleted, Major, YearLevel,
  stringsAsFactors = FALSE
)

# 3. Display the data frame
print(students)
##   StudentID  Name Age CreditsCompleted       Major YearLevel
## 1      S001 Alice  20               45  Data Sains Sophomore
## 2      S002  Budi  21               60 Mathematics    Junior
## 3      S003 Citra  19               30  Statistics  Freshman

2 Exercise 2

Identify Data Types: Determine the type of data for each of the following variables:

# Install knitr package if not already installed
# install.packages("knitr")
library(knitr) #untuk tabel yang rapih 
variables_info <- data.frame( 
  No = 1:5, # data yang di tampilkan 1 hingga 5
  
  #c itu Combine. untuk membuat vector (data)
  Variable = c( 
    "Number of vehicles passing through the toll road each day",
    "Student height in cm",
    "Employee gender (Male / Female)",
    "Customer satisfaction level: Low, Medium, High",
    "Respondent's favorite color: Red, Blue, Green"
  ),
  DataType = c( 
    "Numeric",
    "Numeric",
    "Categorical",
    "Categorical",
    "Categorical"
  ),
  Subtype = c(
    "Continous",
    "Continous",
    "Nominal",
    "Ordinal",
    "Nominal"
  ),
  #Mengatur agar data teks (string) tidak otomatis diubah menjadi factor
  stringsAsFactors = FALSE 
)

# Display the data frame as a neat table
kable(variables_info, 
      caption = "Table of Variables and Data Types") #untuk caption (keterangan) yang muncul di atas tabel
Table of Variables and Data Types
No Variable DataType Subtype
1 Number of vehicles passing through the toll road each day Numeric Continous
2 Student height in cm Numeric Continous
3 Employee gender (Male / Female) Categorical Nominal
4 Customer satisfaction level: Low, Medium, High Categorical Ordinal
5 Respondent’s favorite color: Red, Blue, Green Categorical Nominal

3 Exercise 3

Classify Data Sources: Determine whether the following data comes from internal or external sources, and whether it is structured or unstructured:

# Install DT package if not already installed
# install.packages("DT")
library(DT) # package DT(Data Tables) untuk membuat tabel data yang rapih

data_sources <- data.frame(
  No = 1:4, # baris data yang muncul 4 data
  DataSource = c(
    "Daily sales transaction data of the company",
    "Weather reports from BMKG",
    "Product reviews on social media",
    "Warehouse inventory reports"
  ),
  Internal_External = c(
    "Internal",
    "External",
    "External",
    "Internal"
  ),
  Structured_Unstructured = c(
    "Structured",
    "structured",
    "Unstructured",
    "Structured"
  ),
  
  #Mengatur agar data teks (string) tidak otomatis diubah menjadi factor
  stringsAsFactors = FALSE 
)


datatable(data_sources, 
          caption = "Table of Data Sources", # keterangan untuk di atas tabel
          rownames = FALSE) # hides the index column

4 Exercise 4

Dataset Structure: Consider the following transaction table:

Date Qty Price Product CustomerTier
2025-10-01 2 1000 Laptop High
2025-10-01 5 20 Mouse Medium
2025-10-02 1 1000 Laptop Low
2025-10-02 3 30 Keyboard Medium
2025-10-03 4 50 Mouse Medium
2025-10-03 2 1000 Laptop High
2025-10-04 6 25 Keyboard Low
2025-10-04 1 1000 Laptop High
2025-10-05 3 40 Mouse Low
2025-10-05 5 10 Keyboard Medium

Your Assignment Instructions: Creating a Transactions Table above in R

  1. Create a data frame in R called transactions containing the data above.

  2. Identify which variables are numeric and which are categorical

  3. Calculate total revenue for each transaction by multiplying Qty × Price and add it as a new column Total.

  4. Compute summary statistics:

    • Total quantity sold for each product
    • Total revenue per product
    • Average price per product

  5. Visualize the data:

    • Create a barplot showing total quantity sold per product.
    • Create a pie chart showing the proportion of total revenue per customer tier.

  6. Optional Challenge:

    • Find which date had the highest total revenue.
    • Create a stacked bar chart showing quantity sold per product by customer tier.

Hints: Use data.frame(), aggregate(), barplot(), pie(), and basic arithmetic operations in R.

library(DT)
NO = 1:10
transactions <- data.frame(
  Date = c("2025-10-01", "2025-10-01", "2025-10-02", "2025-10-02",
                   "2025-10-03", "2025-10-03", "2025-10-04", "2025-10-04", 
                   "2025-10-05", "2025-10-05"),
  Qty = c(2, 5, 1, 3, 4, 2, 6, 1, 3, 5),
  Price = c(1000, 20, 1000, 30, 50, 1000, 25, 1000, 40, 10),
  Product = c("Laptop", "Mouse", "Laptop", "Keyboard", "Mouse",
              "Laptop", "Keyboard", "Laptop", "Mouse", "Keyboard"),
  #di atas ini Vector (karena datanya tidak memiliki level (text (string)))
  
  # di bawah ini Factor (karena menyimpan data Ordinal(ada levelnya Low - Medium - High)) bukan text (string)
  CustomerTier = c("High", "Medium", "Low", "Medium", "Medium",
                   "High", "Low", "High", "Low", "Medium"),
  
  stringsAsFactors = FALSE #Mengatur agar data teks (string) tidak otomatis diubah menjadi factor
)

# Atur urutan CustomerTier (Low → Medium → High)
transactions$CustomerTier <- factor( # ($) ambil kolom tertentu dari data frame yang bernama transactions
  transactions$CustomerTier,
  levels = c("Low", "Medium", "High"), # untuk menentukan urutan kategori
  ordered = TRUE # memberitahu R bahwa kategori (levels) di dalam faktor tersebut memiliki urutan yang logis
                 # dengan argumen ordered = TRUE R jadi tahu kalau High lebih tinggi dari Medium dan seterusnya
)

# Tambahkan kolom Total = Qty × Price
transactions$Total <- transactions$Qty * transactions$Price 


# Tampilkan tabel interaktif dengan DT
datatable(
  transactions,
  caption = "Table: Transaction Data with Total Revenue",
  rownames = FALSE,
  )
transactions_summary <- aggregate(
  cbind(Qty, Total, Price) ~ Product, # Untuk menggabungkan isi data (value) pada tabel Qty, Total, Price
  data = transactions,
  
  # ini fungsi (function) untuk data di bawahnya (sum = penjumlahan untuk nilai total)
  # (mean = menghitung nilai rata rata)
  FUN = function(x) c(sum = sum(x), avg = mean(x)) 
)


# Bayangin punya lemari besar namanya transactions_summary, 
# di lemari itu ada  4 laci : 1. Product, 2. Qty, 3. Total, 4. Price
transactions_summary <- data.frame(
  Product = transactions_summary$Product,
  
  # Ambil map 'sum' dari laci 'Qty'
  Total_Qty = transactions_summary$Qty[, "sum"],
  
  # Ambil map 'sum' dari laci 'Total'
  Total_Revenue = transactions_summary$Total[, "sum"],
  
  # Ambil map 'avg' dari laci 'Price', lalu bulatkan ke 2 angka desimal
  Avg_Price = round(transactions_summary$Price[, "avg"], 2)
)


#Tampilkan Table Numeric
numeric_table <- transactions[, c("Date", "Qty", "Price", "Total")] # data yang hanya di ambil Date, Qty, Price, Total

datatable(
  numeric_table,
  caption = "Table 2: Numeric Table",
  rownames = FALSE,
  options = list(pageLength = 5) # hanya menampilkan 5 baris data
)
# Tampilkan Table Categorical
categorical_table <- transactions[, c("Date", "Product", "CustomerTier")] # data yang hanya di ambil Date, Product, CustomerTier

datatable(
  categorical_table,
  caption = "Table 3: Categorical Table",
  rownames = FALSE,
  options = list(pageLength = 5) # hanya menampilkan 5 baris data
)
# Tampilkan tabel Summary
datatable(
  transactions_summary,
  caption = "Table 4: Summary Statistics per Product",
  rownames = FALSE,
  options = list(pageLength = 3) # hanya menampilkan 3 baris data
)
# PEMBUATAN BARPLOT

# aggregate itu untuk Mengelompokkan data. jadi pengelompokan data transactions berdasarkan kolom Product, 
qty_per_product <- aggregate(Qty ~ Product, data = transactions, sum) # Qty ~ Product -> kelompokkan Qty menurut Product
# Buat barplot
barplot(
  qty_per_product$Qty,  # tinggi batang sesuai jumlah penjualan
  names.arg = qty_per_product$Product, # nama batang = nama produk
  col = "skyblue", # menampilkan bar warna menjadi biru (skyblue)
  main = "Total Quantity Sold per Product", # Judul barplot
  xlab = "Product", # menunjukkan data sumbu X (Horizontal)
  ylab = "Total Quantity Sold" # menunjukkan data sumbu Y (Vertikal)
  )

# Mengelompokkan data transactions berdasarkan CustomerTier, lalu menjumlahkan nilai Total untuk setiap level pelanggan.
revenue_per_tier <- aggregate(Total ~ CustomerTier, data = transactions, sum) # Total ~ CustomerTier -> kelompokkan Total menurut CustomerTier
# Buat pie chart
pie(
  revenue_per_tier$Total, # nilai total per tier
  labels = revenue_per_tier$CustomerTier, # label tiap potongan pie
  col = c("red", "yellow", "green"), # menampilkan bar warna merah, kuning, hijau.
  main = "Proportion of Total Revenue per Customer Tier" # Judul Pie chart
)

5 Exercise 5

Create Your Own Data Frame:

Objective: Create a data frame in R with 30 rows containing a mix of data types: continuous, discrete, nominal, and ordinal.

5.1 Instructions

  1. Open RStudio or the R console.

  2. Create a vector for each column in your data frame:

    • Date: 30 dates (can be sequential or random within a month/year)
    • Continuous: numeric values that can take decimal values (e.g., height, weight, temperature)
    • Discrete: numeric values that can only take whole numbers (e.g., number of items, number of vehicles)
    • Nominal: categorical values with no order (e.g., color, gender, city)
    • Ordinal: categorical values with a defined order (e.g., Low, Medium, High; Beginner, Intermediate, Expert)

  3. Combine all vectors into a data frame called my_data.

  4. Check your data frame using head() or View() to ensure it has 30 rows and the columns are correct.

  5. Optional tasks:

    • Summarize each column using summary()
    • Count the frequency of each category for Nominal and Ordinal columns using table()

5.2 Hints

  • Use seq.Date() or as.Date() to generate the Date column.
  • Use runif() or rnorm() for continuous numeric data.
  • Use sample() for discrete, nominal, and ordinal data.
  • Ensure the ordinal vector is created with factor(..., levels = c("Low","Medium","High"), ordered = TRUE) (or similar).
# Install kableExtra package if not already installed
# install.packages("kableExtra")
library(knitr)
library(kableExtra) #library untuk double header di table

# 1. Kolom Date: 30 hari berturut-turut di bulan Oktober 2025
Date <- seq.Date(from = as.Date("2025-10-01"), by = "day", length.out = 30)

# 2. Kolom Weather Temperature
# Menggunakan runif() untuk menghasilkan angka desimal acak antara 15°C dan 35°C
Weather_Temperature <- runif(30, min = 15, max = 35)

# 3. Kolom Number of Green Areas
# Menggunakan sample() untuk menghasilkan angka bulat acak antara 1 - 20
Number_of_Green_Areas <- sample(1:20, 30, replace = TRUE) 

  # 4. Kolom City Name
Cities <- c("New York", "Los Angeles", "Chicago", "Houston", "Phoenix",
            "Philadelphia", "San Antonio", "San Diego", "Dallas", "San Jose")
City_Name <- sample(Cities, 30, replace = TRUE) # untuk memilih nama kota secara acak dari daftar

# 5. Kolom Crime Level
# Menggunakan factor() dengan urutan level dari rendah ke tinggi (Low < Medium < High)
Crime_Level <- factor(
  sample(c("Low", "Medium", "High"), 30, replace = TRUE),
  levels = c("Low", "Medium", "High"),
  ordered = TRUE
)

# Menggabungkan semua kolom menjadi satu data frame bernama my_data
my_data <- data.frame(
  Date = Date,
  Weather_Temperature = Weather_Temperature,
  Number_of_Green_Areas = Number_of_Green_Areas,
  City_Name = City_Name,
  Crime_Level = Crime_Level
)

View(my_data)#melihat semua data

#Untuk menampilkan Kolum Data dan Baris header table
kable(my_data,
        caption = "City Environment & Crime Level Dataset",
        col.names = c("Date", "Air Temperature (°C)", "Number of Green Areas", "City Name", "Crime Level"),
        align = "c") %>%
    add_header_above(c(" " = 1, "Continous" = 1, "Discrete" = 1, "Nominal" = 1, "Ordinal" = 1)
)
City Environment & Crime Level Dataset
Continous
Discrete
Nominal
Ordinal
Date Air Temperature (°C) Number of Green Areas City Name Crime Level
2025-10-01 15.39455 17 New York High
2025-10-02 34.66373 1 Philadelphia Medium
2025-10-03 20.08228 8 Philadelphia Medium
2025-10-04 22.32757 4 Chicago Low
2025-10-05 25.24550 20 Philadelphia Medium
2025-10-06 24.82646 14 Phoenix High
2025-10-07 17.03657 7 Houston Low
2025-10-08 25.22392 6 Phoenix Low
2025-10-09 19.31232 16 San Diego Medium
2025-10-10 23.78695 7 Chicago Low
2025-10-11 28.01679 9 Dallas Medium
2025-10-12 21.58532 12 San Diego High
2025-10-13 26.28265 1 Houston Low
2025-10-14 29.72757 1 Chicago Medium
2025-10-15 17.24652 15 San Antonio Low
2025-10-16 28.57331 16 Chicago Medium
2025-10-17 22.22300 3 New York High
2025-10-18 19.39754 10 Philadelphia Medium
2025-10-19 19.13467 12 San Jose High
2025-10-20 25.80432 20 Phoenix Low
2025-10-21 29.95591 3 Chicago Low
2025-10-22 15.96142 20 Houston Low
2025-10-23 33.21069 17 Philadelphia High
2025-10-24 19.30976 18 Philadelphia Low
2025-10-25 30.62475 19 Los Angeles Low
2025-10-26 23.48873 6 Houston Low
2025-10-27 22.11962 13 San Diego Low
2025-10-28 17.75099 13 Phoenix Medium
2025-10-29 19.15034 5 Phoenix High
2025-10-30 34.90841 9 Dallas Medium
---
title: "Data Exploration"       # Main title of the document
subtitle: "Exercises ~ Week 3"  # Subtitle or topic for week 2
author:
- "Refantanur Hunsul Haqib"
- "Cahaya Medina Semidang"
- "Adinda Maiza Ishfahani"
- "Chandra Rizal Alamsyah"
- "Fityanandra Athar Adyaksa"   # Replace with your full name
date:  "`r format(Sys.Date(), '%B %d, %Y')`" # Auto displays the current date
output:                         # Output section defines the format and layout 
  rmdformats::readthedown:      # https://github.com/juba/rmdformats
    self_contained: true        # Embeds all resources (CSS, JS, images) 
    thumbnails: true            # Displays image thumbnails in the doc
    lightbox: true              # Enables click to enlarge images
    gallery: true               # Groups images into an interactive gallery
    number_sections: true       # Automatically numbers all sections
    lib_dir: libs               # Directory where JavaScript/CSS libraries
    df_print: "paged"           # Displays data frames as interactive paged 
    code_folding: "show"        # Allows folding/unfolding R code blocks 
    code_download: yes          # Adds a button to download all R code
---


<img id="Foto" src="https://raw.githubusercontent.com/YanDraa/Week3Statistika/main/FOTO_KELOMPOK.jpg" alt="Logo" style="width:200px; display: block; margin: auto;">

---

## Exercise 1

The following table shows sample information for three students. Each observation represents a single student and includes details such as their unique student ID, name, age, total credits completed, major field of study, and year level.  

This dataset demonstrates a mixture of variable types:  

- **Nominal:** StudentID, Name, Major  
- **Numeric:** Age (continuous), CreditsCompleted (discrete)  
- **Ordinal:** YearLevel (Freshman → Senior)  

| StudentID | Name   | Age | CreditsCompleted | Major            | YearLevel |
|-----------|--------|-----|-----------------|-----------------|-----------|
| S001      | Alice  | 20  | 45              | Data Sains      | Sophomore |
| S002      | Budi   | 21  | 60              | Mathematics     | Junior    |
| S003      | Citra  | 19  | 30              | Statistics      | Freshman  |

```{r}
# 1. Create vectors for each variable
StudentID <- c("S001", "S002", "S003")       # Nominal / ID
Name <- c("Alice", "Budi", "Citra")          # Nominal / Name
Age <- c(20, 21, 19)                         # Numeric / Continuous
CreditsCompleted <- c(45, 60, 30)            # Numeric / Discrete

# Nominal
Major <- c("Data Sains", "Mathematics", "Statistics")  

# Ordinal
YearLevel <- factor(c("Sophomore", "Junior", "Freshman"),
                    levels = c("Freshman","Sophomore","Junior","Senior"),
                    ordered = TRUE)          

# 2. Combine all vectors into a data frame
students <- data.frame(
  StudentID, Name, Age, CreditsCompleted, Major, YearLevel,
  stringsAsFactors = FALSE
)

# 3. Display the data frame
print(students)
```


## Exercise 2

**Identify Data Types:** Determine the type of data for each of the following variables:

```{r}
# Install knitr package if not already installed
# install.packages("knitr")
library(knitr) #untuk tabel yang rapih 
variables_info <- data.frame( 
  No = 1:5, # data yang di tampilkan 1 hingga 5
  
  #c itu Combine. untuk membuat vector (data)
  Variable = c( 
    "Number of vehicles passing through the toll road each day",
    "Student height in cm",
    "Employee gender (Male / Female)",
    "Customer satisfaction level: Low, Medium, High",
    "Respondent's favorite color: Red, Blue, Green"
  ),
  DataType = c( 
    "Numeric",
    "Numeric",
    "Categorical",
    "Categorical",
    "Categorical"
  ),
  Subtype = c(
    "Continous",
    "Continous",
    "Nominal",
    "Ordinal",
    "Nominal"
  ),
  #Mengatur agar data teks (string) tidak otomatis diubah menjadi factor
  stringsAsFactors = FALSE 
)

# Display the data frame as a neat table
kable(variables_info, 
      caption = "Table of Variables and Data Types") #untuk caption (keterangan) yang muncul di atas tabel

```
---

## Exercise 3

**Classify Data Sources:** Determine whether the following data comes from **internal** or **external sources**, and whether it is **structured** or **unstructured**:

```{r}
# Install DT package if not already installed
# install.packages("DT")
library(DT) # package DT(Data Tables) untuk membuat tabel data yang rapih

data_sources <- data.frame(
  No = 1:4, # baris data yang muncul 4 data
  DataSource = c(
    "Daily sales transaction data of the company",
    "Weather reports from BMKG",
    "Product reviews on social media",
    "Warehouse inventory reports"
  ),
  Internal_External = c(
    "Internal",
    "External",
    "External",
    "Internal"
  ),
  Structured_Unstructured = c(
    "Structured",
    "structured",
    "Unstructured",
    "Structured"
  ),
  
  #Mengatur agar data teks (string) tidak otomatis diubah menjadi factor
  stringsAsFactors = FALSE 
)


datatable(data_sources, 
          caption = "Table of Data Sources", # keterangan untuk di atas tabel
          rownames = FALSE) # hides the index column
```

---

## Exercise 4

**Dataset Structure:** Consider the following transaction table:

| Date       | Qty | Price | Product  | CustomerTier |
|------------|-----|-------|----------|--------------|
| 2025-10-01 | 2   | 1000  | Laptop   | High         |
| 2025-10-01 | 5   | 20    | Mouse    | Medium       |
| 2025-10-02 | 1   | 1000  | Laptop   | Low          |
| 2025-10-02 | 3   | 30    | Keyboard | Medium       |
| 2025-10-03 | 4   | 50    | Mouse    | Medium       |
| 2025-10-03 | 2   | 1000  | Laptop   | High         |
| 2025-10-04 | 6   | 25    | Keyboard | Low          |
| 2025-10-04 | 1   | 1000  | Laptop   | High         |
| 2025-10-05 | 3   | 40    | Mouse    | Low          |
| 2025-10-05 | 5   | 10    | Keyboard | Medium       |


**Your Assignment Instructions:** Creating a Transactions Table above in R

1. **Create a data frame** in R called `transactions` containing the data above.

2. Identify which variables are numeric and which are categorical

3. **Calculate total revenue** for each transaction by multiplying `Qty × Price` and add it as a new column `Total`.

4. **Compute summary statistics**:
   - Total quantity sold for each product
   - Total revenue per product
   - Average price per product

5. **Visualize the data**:
   - Create a **barplot** showing total quantity sold per product.
   - Create a **pie chart** showing the proportion of total revenue per customer tier.

6. **Optional Challenge**:
   - Find which **date** had the highest total revenue.
   - Create a **stacked bar chart** showing quantity sold per product by customer tier.

**Hints:** Use `data.frame()`, `aggregate()`, `barplot()`, `pie()`, and basic arithmetic operations in R.

```{r}
library(DT)
NO = 1:10
transactions <- data.frame(
  Date = c("2025-10-01", "2025-10-01", "2025-10-02", "2025-10-02",
                   "2025-10-03", "2025-10-03", "2025-10-04", "2025-10-04", 
                   "2025-10-05", "2025-10-05"),
  Qty = c(2, 5, 1, 3, 4, 2, 6, 1, 3, 5),
  Price = c(1000, 20, 1000, 30, 50, 1000, 25, 1000, 40, 10),
  Product = c("Laptop", "Mouse", "Laptop", "Keyboard", "Mouse",
              "Laptop", "Keyboard", "Laptop", "Mouse", "Keyboard"),
  #di atas ini Vector (karena datanya tidak memiliki level (text (string)))
  
  # di bawah ini Factor (karena menyimpan data Ordinal(ada levelnya Low - Medium - High)) bukan text (string)
  CustomerTier = c("High", "Medium", "Low", "Medium", "Medium",
                   "High", "Low", "High", "Low", "Medium"),
  
  stringsAsFactors = FALSE #Mengatur agar data teks (string) tidak otomatis diubah menjadi factor
)

# Atur urutan CustomerTier (Low → Medium → High)
transactions$CustomerTier <- factor( # ($) ambil kolom tertentu dari data frame yang bernama transactions
  transactions$CustomerTier,
  levels = c("Low", "Medium", "High"), # untuk menentukan urutan kategori
  ordered = TRUE # memberitahu R bahwa kategori (levels) di dalam faktor tersebut memiliki urutan yang logis
                 # dengan argumen ordered = TRUE R jadi tahu kalau High lebih tinggi dari Medium dan seterusnya
)

# Tambahkan kolom Total = Qty × Price
transactions$Total <- transactions$Qty * transactions$Price 


# Tampilkan tabel interaktif dengan DT
datatable(
  transactions,
  caption = "Table: Transaction Data with Total Revenue",
  rownames = FALSE,
  )

transactions_summary <- aggregate(
  cbind(Qty, Total, Price) ~ Product, # Untuk menggabungkan isi data (value) pada tabel Qty, Total, Price
  data = transactions,
  
  # ini fungsi (function) untuk data di bawahnya (sum = penjumlahan untuk nilai total)
  # (mean = menghitung nilai rata rata)
  FUN = function(x) c(sum = sum(x), avg = mean(x)) 
)


# Bayangin punya lemari besar namanya transactions_summary, 
# di lemari itu ada  4 laci : 1. Product, 2. Qty, 3. Total, 4. Price
transactions_summary <- data.frame(
  Product = transactions_summary$Product,
  
  # Ambil map 'sum' dari laci 'Qty'
  Total_Qty = transactions_summary$Qty[, "sum"],
  
  # Ambil map 'sum' dari laci 'Total'
  Total_Revenue = transactions_summary$Total[, "sum"],
  
  # Ambil map 'avg' dari laci 'Price', lalu bulatkan ke 2 angka desimal
  Avg_Price = round(transactions_summary$Price[, "avg"], 2)
)


#Tampilkan Table Numeric
numeric_table <- transactions[, c("Date", "Qty", "Price", "Total")] # data yang hanya di ambil Date, Qty, Price, Total

datatable(
  numeric_table,
  caption = "Table 2: Numeric Table",
  rownames = FALSE,
  options = list(pageLength = 5) # hanya menampilkan 5 baris data
)

# Tampilkan Table Categorical
categorical_table <- transactions[, c("Date", "Product", "CustomerTier")] # data yang hanya di ambil Date, Product, CustomerTier

datatable(
  categorical_table,
  caption = "Table 3: Categorical Table",
  rownames = FALSE,
  options = list(pageLength = 5) # hanya menampilkan 5 baris data
)

# Tampilkan tabel Summary
datatable(
  transactions_summary,
  caption = "Table 4: Summary Statistics per Product",
  rownames = FALSE,
  options = list(pageLength = 3) # hanya menampilkan 3 baris data
)


# PEMBUATAN BARPLOT

# aggregate itu untuk Mengelompokkan data. jadi pengelompokan data transactions berdasarkan kolom Product, 
qty_per_product <- aggregate(Qty ~ Product, data = transactions, sum) # Qty ~ Product -> kelompokkan Qty menurut Product
# Buat barplot
barplot(
  qty_per_product$Qty,  # tinggi batang sesuai jumlah penjualan
  names.arg = qty_per_product$Product, # nama batang = nama produk
  col = "skyblue", # menampilkan bar warna menjadi biru (skyblue)
  main = "Total Quantity Sold per Product", # Judul barplot
  xlab = "Product", # menunjukkan data sumbu X (Horizontal)
  ylab = "Total Quantity Sold" # menunjukkan data sumbu Y (Vertikal)
  )

# Mengelompokkan data transactions berdasarkan CustomerTier, lalu menjumlahkan nilai Total untuk setiap level pelanggan.
revenue_per_tier <- aggregate(Total ~ CustomerTier, data = transactions, sum) # Total ~ CustomerTier -> kelompokkan Total menurut CustomerTier
# Buat pie chart
pie(
  revenue_per_tier$Total, # nilai total per tier
  labels = revenue_per_tier$CustomerTier, # label tiap potongan pie
  col = c("red", "yellow", "green"), # menampilkan bar warna merah, kuning, hijau.
  main = "Proportion of Total Revenue per Customer Tier" # Judul Pie chart
)

```

## Exercise 5

**Create Your Own Data Frame:**

**Objective:** Create a data frame in R with **30 rows** containing a mix of data types: continuous, discrete, nominal, and ordinal.  

### Instructions

1. **Open RStudio** or the R console.  

2. **Create a vector for each column** in your data frame:  

   - **Date**: 30 dates (can be sequential or random within a month/year)  
   - **Continuous**: numeric values that can take decimal values (e.g., height, weight, temperature)  
   - **Discrete**: numeric values that can only take whole numbers (e.g., number of items, number of vehicles)  
   - **Nominal**: categorical values with **no order** (e.g., color, gender, city)  
   - **Ordinal**: categorical values with a **defined order** (e.g., Low, Medium, High; Beginner, Intermediate, Expert)  

3. **Combine all vectors into a data frame** called `my_data`.  

4. **Check your data frame** using `head()` or `View()` to ensure it has **30 rows** and the columns are correct.  

5. **Optional tasks**:  
   - Summarize each column using `summary()`  
   - Count the frequency of each category for **Nominal** and **Ordinal** columns using `table()`  

### Hints

- Use `seq.Date()` or `as.Date()` to generate the Date column.  
- Use `runif()` or `rnorm()` for continuous numeric data.  
- Use `sample()` for discrete, nominal, and ordinal data.  
- Ensure the **ordinal vector** is created with `factor(..., levels = c("Low","Medium","High"), ordered = TRUE)` (or similar).  

```{r}
# Install kableExtra package if not already installed
# install.packages("kableExtra")
library(knitr)
library(kableExtra) #library untuk double header di table

# 1. Kolom Date: 30 hari berturut-turut di bulan Oktober 2025
Date <- seq.Date(from = as.Date("2025-10-01"), by = "day", length.out = 30)

# 2. Kolom Weather Temperature
# Menggunakan runif() untuk menghasilkan angka desimal acak antara 15°C dan 35°C
Weather_Temperature <- runif(30, min = 15, max = 35)

# 3. Kolom Number of Green Areas
# Menggunakan sample() untuk menghasilkan angka bulat acak antara 1 - 20
Number_of_Green_Areas <- sample(1:20, 30, replace = TRUE) 

  # 4. Kolom City Name
Cities <- c("New York", "Los Angeles", "Chicago", "Houston", "Phoenix",
            "Philadelphia", "San Antonio", "San Diego", "Dallas", "San Jose")
City_Name <- sample(Cities, 30, replace = TRUE) # untuk memilih nama kota secara acak dari daftar

# 5. Kolom Crime Level
# Menggunakan factor() dengan urutan level dari rendah ke tinggi (Low < Medium < High)
Crime_Level <- factor(
  sample(c("Low", "Medium", "High"), 30, replace = TRUE),
  levels = c("Low", "Medium", "High"),
  ordered = TRUE
)

# Menggabungkan semua kolom menjadi satu data frame bernama my_data
my_data <- data.frame(
  Date = Date,
  Weather_Temperature = Weather_Temperature,
  Number_of_Green_Areas = Number_of_Green_Areas,
  City_Name = City_Name,
  Crime_Level = Crime_Level
)

View(my_data)#melihat semua data

#Untuk menampilkan Kolum Data dan Baris header table
kable(my_data,
        caption = "City Environment & Crime Level Dataset",
        col.names = c("Date", "Air Temperature (°C)", "Number of Green Areas", "City Name", "Crime Level"),
        align = "c") %>%
    add_header_above(c(" " = 1, "Continous" = 1, "Discrete" = 1, "Nominal" = 1, "Ordinal" = 1)
) 

```



