Functions Loops Practicum

Assignment Week 5

Fityanandra Athar Adyaksa (52250059)

Data Science students at

Enthusiastic about learning

April 07, 2026

---

Preface

This practicum report is a consolidated document covering Task 1 through Task 8 of the Data Science Programming — Functions and Loops module. All tasks are unified into a single file to provide a coherent, end-to-end reference that traces the progressive application of core programming concepts across increasingly complex analytical scenarios.

The practicum centres on three fundamental pillars of programming: functions, loops, and conditional branching. These concepts underpin virtually all data wrangling, simulation, and automated reporting workflows in professional data science practice. Each task is implemented in both R and Python to reinforce language-agnostic thinking and to highlight syntactic differences between the two ecosystems.

All visualisations in this document are interactive, rendered via the plotly library in R and plotly in Python, allowing readers to hover for precise values, zoom into regions of interest, and toggle series visibility directly in the browser.

---

Research Objectives

The overarching objectives of this practicum are:

1. Understand and implement functions in R and Python as reusable, self-contained units of logic — including input validation and safe error handling.
2. Master loop constructs — both single and nested for loops — to automate repetitive computation across multi-dimensional datasets.
3. Apply conditional logic (if-else if-else / if-elif-else) for data-driven decision making: tiered discounts, performance classification, and KPI assignment.
4. Build and evaluate data simulations using Monte Carlo sampling and multi-entity nested loop structures.
5. Perform data transformation and feature engineering through min-max normalisation, Z-score standardisation, and categorical variable creation.
6. Communicate results through interactive visualisation using plotly for hover-enabled, zoomable charts.
7. Automate report generation by encapsulating all reporting logic inside a function called iteratively via a loop — the foundation of production-grade automated reporting pipelines.

---

Task 1 — Function Definition, Nested Loops & Formula Comparison

This task covers Section 3 of Data Science Programming — Functions and Loops. The specific objectives are:

1. Build a dynamic multi-formula function with conditional branching to evaluate four mathematical expressions.
2. Apply a nested loop to compute formula values across x = 1 to 20.
3. Implement input validation so the function fails safely on unrecognised inputs.
4. Produce an interactive comparative visualisation of all four formulas.

Formulas Overview

Formula Type	Expression	Behaviour
Linear	f(x) = 2x + 3	Constant growth rate — straight line
Quadratic	f(x) = x² - 2x + 1	Accelerating growth — U-shaped curve
Cubic	f(x) = 0.5x³ - 3x + 2	Initial dip, then steep rise
Exponential	f(x) = 2e^(0.2x)	Fastest growth — compounds repeatedly

---

Task 1.1 — Function Definition & Input Validation

compute_formula() accepts a numeric x and a string formula, returning the computed result. The else branch acts as an input guard, returning NA / None with a descriptive warning rather than crashing silently.

Feature	R	Python
Branch keyword	else if	elif
Exponential	exp()	math.exp()
Invalid input handling	warning() + return(NA)	print() + return None

— R

# compute_formula: evaluates one of four mathematical formulas at a given x.
# Returns NA with a warning if an unrecognised formula name is supplied.
compute_formula <- function(x, formula) {
  if      (formula == "linear")      return(2 * x + 3)
  else if (formula == "quadratic")   return(x^2 - 2 * x + 1)
  else if (formula == "cubic")       return(0.5 * x^3 - 3 * x + 2)
  else if (formula == "exponential") return(2 * exp(0.2 * x))
  else {
    warning(paste0("[compute_formula] Unknown formula: '", formula, "'."))
    return(NA)
  }
}

cat("=== Spot Checks (x = 5) ===\n")

## === Spot Checks (x = 5) ===

cat("linear      :", compute_formula(5, "linear"),                "\n")

## linear      : 13

cat("quadratic   :", compute_formula(5, "quadratic"),             "\n")

## quadratic   : 16

cat("cubic       :", compute_formula(5, "cubic"),                 "\n")

## cubic       : 49.5

cat("exponential :", round(compute_formula(5, "exponential"), 4), "\n")

## exponential : 5.4366

cat("\n=== Validation — Unknown Formula ===\n")

## 
## === Validation — Unknown Formula ===

cat("Returned   :", compute_formula(5, "logarithmic"), "\n")

## Returned   : NA

---

Task 1.2 — Nested Loop & Dataset Construction

A nested loop evaluates every formula at every x from 1 to 20, producing an 80-row dataset (4 formulas × 20 values).

— R

formula_list <- c("linear", "quadratic", "cubic", "exponential")
x_range      <- 1:20

# FIX: gunakan list lalu do.call(rbind) — jauh lebih cepat dari rbind() per iterasi
rows_list <- vector("list", length(formula_list) * length(x_range))
idx <- 1L
for (f in formula_list) {
  for (x in x_range) {
    rows_list[[idx]] <- data.frame(x = x, y = compute_formula(x, f),
                                   formula = f, stringsAsFactors = FALSE)
    idx <- idx + 1L
  }
}
results_df <- do.call(rbind, rows_list)

cat("Rows generated:", nrow(results_df),
    "(", length(formula_list), "formulas ×", length(x_range), "x values )\n\n")

## Rows generated: 80 ( 4 formulas × 20 x values )

head(results_df, 8)

---

Task 1.3 — Interactive Visualisation: Four Formulas on One Plot

— R

formula_colours <- c(
  "linear"      = "#2196F3",
  "quadratic"   = "#4CAF50",
  "cubic"       = "#FF9800",
  "exponential" = "#E91E63"
)
formula_labels <- c(
  "linear"      = "Linear: f(x) = 2x + 3",
  "quadratic"   = "Quadratic: f(x) = x\u00B2 - 2x + 1",
  "cubic"       = "Cubic: f(x) = 0.5x\u00B3 - 3x + 2",
  "exponential" = "Exponential: f(x) = 2e^(0.2x)"
)

p1 <- plot_ly()
for (f in formula_list) {
  df_sub <- results_df[results_df$formula == f, ]
  p1 <- add_trace(p1,
    data      = df_sub, x = ~x, y = ~y,
    type      = "scatter", mode = "lines+markers",
    name      = formula_labels[[f]],
    line      = list(color = formula_colours[[f]], width = 2.5),
    marker    = list(color = formula_colours[[f]], size = 6))
}
p1 <- layout(p1,
  title     = list(text = "Task 1: Comparison of Four Mathematical Formulas (x = 1 to 20)",
                   font = list(size = 15)),
  xaxis     = list(title = "x value", showgrid = TRUE, gridcolor = "#e8e8e8"),
  yaxis     = list(title = "f(x)",    showgrid = TRUE, gridcolor = "#e8e8e8"),
  legend    = list(title = list(text = "<b>Formula</b>")),
  hovermode = "x unified",
  plot_bgcolor  = "white",
  paper_bgcolor = "white")
p1

---

Task 1.4 — Descriptive Statistics per Formula

— R

summary_table <- results_df %>%
  group_by(formula) %>%
  summarise(
    Min    = round(min(y),    2),
    Max    = round(max(y),    2),
    Mean   = round(mean(y),   2),
    Median = round(median(y), 2),
    SD     = round(sd(y),     2),
    .groups = "drop"
  ) %>% rename(Formula = formula)

kable(summary_table,
      col.names = c("Formula","Min","Max","Mean","Median","Std Dev"),
      align = "lrrrrr",
      caption = "Descriptive statistics for each formula across x = 1 to 20")

Descriptive statistics for each formula across x = 1 to 20

Formula	Min	Max	Mean	Median	Std Dev
cubic	-0.50	3942.0	1073.00	553.25	1236.10
exponential	2.44	109.2	29.57	16.41	31.35
linear	5.00	43.0	24.00	24.00	11.83
quadratic	0.00	361.0	123.50	90.50	116.44

---

Summary — Task 1

• Function design with conditional branching provides a clean, reusable interface for four distinct formula types.
• Nested loops are the natural structure for two-dimensional computation, producing an 80-row dataset with minimal code.
• Input validation via the else branch ensures safe failure — the function returns NA/None with a descriptive warning.
• Visualisation confirms that exponential growth dominates at large x, while linear remains the most predictable throughout.

---

Task 2 — Nested Sales Simulation with Conditional Discounts

The objectives for Task 2 are:

1. Build simulate_sales() — a nested simulation generating a multi-salesperson, multi-day sales dataset.
2. Apply tiered conditional discount logic based on sales amount thresholds.
3. Compute cumulative sales per salesperson using a nested helper function.
4. Produce an interactive cumulative sales chart.

Simulation Parameters

Parameter / Rule	Detail
n_salesperson	Number of salespersons to simulate
days	Number of trading days per salesperson
Discount — High	sales_amount > 800 → 20% discount
Discount — Medium	sales_amount > 500 → 10% discount
No Discount	sales_amount ≤ 500 → no discount

---

Task 2.1 — Sales Simulation Function

— R

set.seed(42)

# simulate_sales: generates a sales dataset with tiered discount logic.
# FIX: gunakan list lalu do.call(rbind) — jauh lebih cepat dari rbind() per iterasi
simulate_sales <- function(n_salesperson, days) {
  rows_list <- vector("list", n_salesperson * days)
  idx <- 1L
  for (s in 1:n_salesperson) {
    for (d in 1:days) {
      amount   <- round(runif(1, 200, 1000), 2)
      discount <- if      (amount > 800) 0.20
                  else if (amount > 500) 0.10
                  else                  0.00
      rows_list[[idx]] <- data.frame(sales_id = s, day = d,
                                     sales_amount = amount, discount_rate = discount,
                                     stringsAsFactors = FALSE)
      idx <- idx + 1L
    }
  }
  do.call(rbind, rows_list)
}

sales_data <- simulate_sales(n_salesperson = 5, days = 10)
cat("Dataset:", nrow(sales_data), "rows ×", ncol(sales_data), "columns\n\n")

## Dataset: 50 rows × 4 columns

head(sales_data, 10)

---

Task 2.2 — Cumulative Sales per Salesperson

— R

calc_cumulative <- function(df, sid) {
  sub_df <- df[df$sales_id == sid, ]
  sub_df <- sub_df[order(sub_df$day), ]
  cumsum(sub_df$sales_amount)
}

# FIX: gunakan list lalu do.call(rbind)
cum_rows <- vector("list", length(unique(sales_data$sales_id)))
for (i in seq_along(unique(sales_data$sales_id))) {
  s <- sort(unique(sales_data$sales_id))[i]
  cum_vals <- calc_cumulative(sales_data, s)
  cum_rows[[i]] <- data.frame(sales_id = s, day = seq_along(cum_vals),
                               cumulative = cum_vals, stringsAsFactors = FALSE)
}
cum_df <- do.call(rbind, cum_rows)

head(cum_df, 10)

---

Task 2.3 — Summary Statistics & Interactive Cumulative Sales Plot

— R

summary_sales <- sales_data %>%
  group_by(sales_id) %>%
  summarise(
    Total_Sales    = round(sum(sales_amount), 2),
    Avg_Daily      = round(mean(sales_amount), 2),
    Total_Discount = round(sum(sales_amount * discount_rate), 2),
    Days_High      = sum(discount_rate == 0.20),
    Days_Medium    = sum(discount_rate == 0.10),
    Days_None      = sum(discount_rate == 0.00),
    .groups = "drop"
  ) %>% rename("Salesperson ID" = sales_id)

kable(summary_sales,
      col.names = c("Salesperson","Total Sales","Avg Daily",
                    "Total Discount","Days High","Days Medium","Days None"),
      align = "crrrrrr",
      caption = "Summary statistics per salesperson (5 persons × 10 days)")

Summary statistics per salesperson (5 persons × 10 days)

Salesperson	Total Sales	Avg Daily	Total Discount	Days High	Days Medium	Days None
1	7090.09	709.01	909.93	3	5	2
2	6720.24	672.02	890.42	3	5	2
3	6923.09	692.31	1101.12	5	3	2
4	6153.74	615.37	801.90	3	4	3
5	7066.52	706.65	1066.84	4	5	1

pal2 <- brewer.pal(5, "Set1")
p2   <- plot_ly()
for (i in seq_along(unique(cum_df$sales_id))) {
  s      <- sort(unique(cum_df$sales_id))[i]
  df_sub <- cum_df[cum_df$sales_id == s, ]
  p2 <- add_trace(p2,
    data = df_sub, x = ~day, y = ~cumulative,
    type = "scatter", mode = "lines+markers",
    name = paste("Salesperson", s),
    line   = list(color = pal2[i], width = 2.2),
    marker = list(color = pal2[i], size  = 6))
}
p2 <- layout(p2,
  title     = list(text = "Task 2: Cumulative Sales per Salesperson (10 Days)"),
  xaxis     = list(title = "Day", dtick = 1),
  yaxis     = list(title = "Cumulative Sales Amount"),
  hovermode = "x unified",
  legend    = list(title = list(text = "<b>Salesperson</b>")),
  plot_bgcolor  = "white",
  paper_bgcolor = "white")
p2

---

Task 3 — Multi-Level Performance Categorisation

The objectives for Task 3 are:

1. Build categorize_performance() — a function assigning one of five performance labels to a sales amount.
2. Apply the function across a vector of 100 values using a loop.
3. Compute the percentage distribution across all five categories.
4. Visualise with an interactive bar chart and pie chart.

Performance Categories

Category	Condition	Plot Colour
Excellent	sales_amount > 800	#2ecc71
Very Good	sales_amount > 650	#3498db
Good	sales_amount > 500	#f39c12
Average	sales_amount > 350	#e67e22
Poor	sales_amount ≤ 350	#e74c3c

---

Task 3.1 — Five-Level Classification Function

— R

set.seed(7)

categorize_performance <- function(sales_amount) {
  if      (sales_amount > 800) return("Excellent")
  else if (sales_amount > 650) return("Very Good")
  else if (sales_amount > 500) return("Good")
  else if (sales_amount > 350) return("Average")
  else                          return("Poor")
}

sales_vector <- round(runif(100, 100, 1000), 2)
# FIX: pre-alokasi vektor karakter, bukan rbind row per row
categories   <- character(length(sales_vector))
for (i in seq_along(sales_vector)) {
  categories[i] <- categorize_performance(sales_vector[i])
}

result_df <- data.frame(sales_amount = sales_vector, category = categories,
                         stringsAsFactors = FALSE)
head(result_df, 10)

---

Task 3.2 — Category Frequency Distribution

— R

level_order           <- c("Excellent","Very Good","Good","Average","Poor")
freq_table            <- as.data.frame(table(result_df$category))
colnames(freq_table)  <- c("Category","Count")
freq_table$Percentage <- round(freq_table$Count / nrow(result_df) * 100, 1)
freq_table$Category   <- factor(freq_table$Category, levels = level_order)
freq_table            <- freq_table[order(freq_table$Category), ]

kable(freq_table, col.names = c("Category","Count","Percentage (%)"),
      align = "lrr",
      caption = "Distribution of performance categories across 100 sales values")

Distribution of performance categories across 100 sales values

	Category	Count	Percentage (%)
2	Excellent	22	22
5	Very Good	17	17
3	Good	15	15
1	Average	21	21
4	Poor	25	25

---

Task 3.3 — Interactive Bar Chart & Pie Chart

— R

cat_colours  <- c("Excellent"="#2ecc71","Very Good"="#3498db",
                  "Good"="#f39c12","Average"="#e67e22","Poor"="#e74c3c")
freq_plot    <- freq_table
freq_plot$Category <- as.character(freq_plot$Category)

p_bar3 <- plot_ly(freq_plot,
  x      = ~Category, y = ~Count,
  type   = "bar",
  marker = list(color = unname(cat_colours[freq_plot$Category])),
  text   = ~paste0(Percentage, "%"),
  textposition = "outside",
  hovertemplate = "<b>%{x}</b><br>Count: %{y}<br>Share: %{text}<extra></extra>"
) %>% layout(
  title  = "Task 3: Performance Category Distribution (n = 100)",
  xaxis  = list(title = "Category", categoryorder = "array",
                categoryarray = level_order),
  yaxis  = list(title = "Count"),
  showlegend   = FALSE,
  plot_bgcolor  = "white",
  paper_bgcolor = "white")

p_pie3 <- plot_ly(freq_plot,
  labels  = ~Category, values = ~Count,
  type    = "pie",
  marker  = list(colors = unname(cat_colours[freq_plot$Category])),
  textinfo      = "label+percent",
  hovertemplate = "<b>%{label}</b><br>Count: %{value}<br>%{percent}<extra></extra>"
) %>% layout(
  title         = "Task 3: Proportional Share by Category",
  plot_bgcolor  = "white",
  paper_bgcolor = "white")

p_bar3

p_pie3

---

Task 4 — Multi-Company Dataset Simulation

The objectives for Task 4 are:

1. Build generate_company_data() with a nested loop to simulate employee records across multiple companies.
2. Apply conditional KPI assignment for top performers (performance score > 75).
3. Produce a per-company summary table.
4. Visualise with an interactive grouped bar chart and scatter plot.

Dataset Schema

Column	Type	Description
company_id	integer	Company identifier (1 to n_company)
employee_id	integer	Employee identifier (1 to n_employees per company)
salary	numeric	Monthly salary in thousands — Uniform(3, 15)
department	character	One of: HR, IT, Finance, Marketing, Operations
performance_score	numeric	Performance score — Normal(70, 15), capped to [0, 100]
KPI_score	numeric	KPI: Uniform(91,100) if top performer, else Uniform(50,89)

---

Task 4.1 — Company Data Generation Function

— R

set.seed(123)

# FIX: gunakan list lalu do.call(rbind) — jauh lebih cepat dari rbind() per iterasi
generate_company_data <- function(n_company, n_employees) {
  departments <- c("HR","IT","Finance","Marketing","Operations")
  rows_list   <- vector("list", n_company * n_employees)
  idx <- 1L
  for (c in 1:n_company) {
    for (e in 1:n_employees) {
      salary <- round(runif(1, 3, 15), 2)
      dept   <- sample(departments, 1)
      perf   <- round(max(0, min(100, rnorm(1, mean = 70, sd = 15))), 1)
      kpi    <- if (perf > 75) round(runif(1, 91, 100), 1) else round(runif(1, 50, 89), 1)
      rows_list[[idx]] <- data.frame(company_id = c, employee_id = e, salary = salary,
                                     department = dept, performance_score = perf,
                                     KPI_score = kpi, stringsAsFactors = FALSE)
      idx <- idx + 1L
    }
  }
  do.call(rbind, rows_list)
}

company_data <- generate_company_data(n_company = 4, n_employees = 20)
cat("Dataset:", nrow(company_data), "rows ×", ncol(company_data), "columns\n\n")

## Dataset: 80 rows × 6 columns

head(company_data, 8)

---

Task 4.2 — Per-Company Summary Table

— R

company_summary <- company_data %>%
  group_by(company_id) %>%
  summarise(
    Avg_Salary      = round(mean(salary), 2),
    Avg_Performance = round(mean(performance_score), 1),
    Max_KPI         = round(max(KPI_score), 1),
    Top_Performers  = sum(performance_score > 75),
    .groups = "drop"
  ) %>% rename("Company ID" = company_id)

kable(company_summary,
      col.names = c("Company","Avg Salary (k)","Avg Performance","Max KPI","Top Performers"),
      align = "crrrc",
      caption = "Per-company summary: 4 companies × 20 employees each")

Per-company summary: 4 companies × 20 employees each

Company	Avg Salary (k)	Avg Performance	Max KPI	Top Performers
1	9.31	73.3	97.4	8
2	9.29	71.0	99.4	7
3	9.92	67.8	99.6	5
4	8.71	71.8	99.2	6

---

Task 4.3 — Interactive Grouped Bar Chart & Scatter Plot

— R

bar_data <- company_summary %>%
  rename(company_id = "Company ID") %>%
  select(company_id, Avg_Salary, Avg_Performance) %>%
  pivot_longer(cols = c(Avg_Salary, Avg_Performance),
               names_to = "Metric", values_to = "Value")

p_bar4 <- plot_ly(bar_data,
  x      = ~factor(company_id), y = ~Value,
  color  = ~Metric,
  colors = c("Avg_Salary" = "#3498db","Avg_Performance" = "#e74c3c"),
  type   = "bar",
  text   = ~round(Value, 2), textposition = "outside"
) %>% layout(
  barmode = "group",
  title   = "Task 4: Average Salary vs. Performance by Company",
  xaxis   = list(title = "Company ID"),
  yaxis   = list(title = "Value"),
  legend  = list(title = list(text = "<b>Metric</b>")),
  plot_bgcolor  = "white",
  paper_bgcolor = "white")

company_data$company_label <- paste("Company", company_data$company_id)

p_scatter4 <- plot_ly(company_data,
  x      = ~performance_score, y = ~KPI_score,
  color  = ~company_label,
  type   = "scatter", mode = "markers",
  marker = list(size = 8, opacity = 0.7),
  hovertemplate = "<b>%{color}</b><br>Performance: %{x}<br>KPI: %{y}<extra></extra>"
) %>% layout(
  title  = "Task 4: Performance Score vs. KPI Score (per Employee)",
  xaxis  = list(title = "Performance Score"),
  yaxis  = list(title = "KPI Score"),
  legend = list(title = list(text = "<b>Company</b>")),
  plot_bgcolor  = "white",
  paper_bgcolor = "white")

p_bar4

p_scatter4

---

Task 5 — Monte Carlo Simulation: π Estimation & Probability

The objectives for Task 5 are:

1. Build monte_carlo_pi() to estimate π by random point sampling in a unit square.
2. Use a loop to count points falling inside the unit circle.
3. Compute the probability of a point landing in a sub-square region.
4. Visualise with an interactive scatter plot.

The Monte Carlo Principle

Item	Description
Unit square	x ∈ [0,1], y ∈ [0,1] — area = 1
Quarter-circle	x² + y² ≤ 1 — area = π/4
Key ratio	points inside circle / total ≈ π/4
π estimate	π ≈ 4 × (points inside / total points)
Sub-square	x ∈ [0,0.5], y ∈ [0,0.5] — area = 0.25 → P ≈ 0.25

---

Task 5.1 — Monte Carlo π Estimation

— R

set.seed(2024)

# FIX: pre-alokasi vektor numerik, bukan rbind per iterasi
monte_carlo_pi <- function(n_points) {
  inside <- 0L; outside <- 0L
  # Pre-alokasi vektor untuk plotting (max 2000 titik)
  plot_n  <- min(n_points, 2000L)
  pts_x   <- numeric(plot_n)
  pts_y   <- numeric(plot_n)
  pts_st  <- character(plot_n)

  for (i in seq_len(n_points)) {
    x <- runif(1); y <- runif(1)
    if (x^2 + y^2 <= 1) { inside  <- inside  + 1L; status <- "inside"  }
    else                 { outside <- outside + 1L; status <- "outside" }
    if (i <= plot_n) {
      pts_x[i] <- x; pts_y[i] <- y; pts_st[i] <- status
    }
  }
  pts <- data.frame(x = pts_x, y = pts_y, status = pts_st, stringsAsFactors = FALSE)
  list(pi_estimate   = 4 * inside / n_points,
       inside_count  = inside,
       outside_count = outside,
       points        = pts)
}

mc_result <- monte_carlo_pi(10000)

cat("=== Monte Carlo π Estimation (n = 10,000) ===\n")

## === Monte Carlo π Estimation (n = 10,000) ===

cat("Points inside circle :", mc_result$inside_count,                      "\n")

## Points inside circle : 7865

cat("Points outside circle:", mc_result$outside_count,                     "\n")

## Points outside circle: 2135

cat("Estimated π          :", round(mc_result$pi_estimate, 6),              "\n")

## Estimated π          : 3.146

cat("True π               :", round(pi, 6),                                "\n")

## True π               : 3.141593

cat("Absolute error       :", round(abs(mc_result$pi_estimate - pi), 6),   "\n")

## Absolute error       : 0.004407

---

Task 5.2 — Sub-Square Probability Analysis

— R

set.seed(2024)
all_x <- runif(10000); all_y <- runif(10000)

in_subsquare   <- sum(all_x <= 0.5 & all_y <= 0.5)
prob_subsquare <- round(in_subsquare / 10000, 4)

cat("=== Sub-Square Probability (x ≤ 0.5 AND y ≤ 0.5) ===\n")

## === Sub-Square Probability (x ≤ 0.5 AND y ≤ 0.5) ===

cat("Points in sub-square  :", in_subsquare,                           "\n")

## Points in sub-square  : 2549

cat("Estimated probability :", prob_subsquare,                         "\n")

## Estimated probability : 0.2549

cat("Theoretical P         : 0.2500\n")

## Theoretical P         : 0.2500

cat("Absolute error        :", round(abs(prob_subsquare - 0.25), 4),  "\n")

## Absolute error        : 0.0049

---

Task 5.3 — Interactive Scatter Plot: Inside vs. Outside the Circle

— R

points_df <- mc_result$points
theta_seq <- seq(0, pi / 2, length.out = 200)
arc_df    <- data.frame(x = cos(theta_seq), y = sin(theta_seq))

p_mc <- plot_ly() %>%
  add_trace(data = points_df[points_df$status == "outside", ],
    x = ~x, y = ~y, type = "scatter", mode = "markers",
    name   = "Outside circle",
    marker = list(color = "#e74c3c", size = 3, opacity = 0.5)) %>%
  add_trace(data = points_df[points_df$status == "inside", ],
    x = ~x, y = ~y, type = "scatter", mode = "markers",
    name   = "Inside circle",
    marker = list(color = "#3498db", size = 3, opacity = 0.5)) %>%
  add_trace(data = arc_df, x = ~x, y = ~y,
    type = "scatter", mode = "lines", name = "Circle boundary",
    line = list(color = "black", width = 1.8)) %>%
  layout(
    title = list(text = paste0(
      "Task 5: Monte Carlo Simulation (2,000 points shown)<br>",
      "<sub>Estimated π = ", round(mc_result$pi_estimate, 5),
      " | True π = ", round(pi, 5), "</sub>")),
    xaxis   = list(title = "x", range = c(0, 1), scaleanchor = "y"),
    yaxis   = list(title = "y", range = c(0, 1)),
    shapes  = list(list(
      type = "rect", x0 = 0, x1 = 0.5, y0 = 0, y1 = 0.5,
      line = list(color = "purple", dash = "dash", width = 1.5),
      fillcolor = "rgba(128,0,128,0.05)")),
    annotations = list(list(
      x = 0.25, y = 0.55, text = "Sub-square (P ≈ 0.25)",
      showarrow = FALSE, font = list(color = "purple", size = 11))),
    plot_bgcolor  = "white",
    paper_bgcolor = "white")
p_mc

---

Task 6 — Advanced Data Transformation & Feature Engineering

The objectives for Task 6 are:

1. Build normalize_columns() — loop-based min-max normalisation (scales to [0, 1]).
2. Build z_score() — loop-based Z-score standardisation (mean = 0, sd = 1).
3. Engineer two categorical features: performance_category and salary_bracket.
4. Compare distributions before and after transformation with interactive histograms.

Transformation Reference

Method	Formula	Output Range	Typical Use Case
Min-Max Normalisation	x_norm = (x − min) / (max − min)	[0, 1] — bounded	Distance-based models, neural networks
Z-Score Standardisation	x_z = (x − mean) / sd	Unbounded — centred at 0	Regression, clustering — unit variance

---

Task 6.1 — Normalisation Functions

— R

set.seed(99)

# FIX: generate_company_data_t6 juga menggunakan list + do.call(rbind)
generate_company_data_t6 <- function(n_company, n_employees) {
  departments <- c("HR","IT","Finance","Marketing","Operations")
  rows_list   <- vector("list", n_company * n_employees)
  idx <- 1L
  for (c in 1:n_company) {
    for (e in 1:n_employees) {
      salary <- round(runif(1, 3, 15), 2)
      dept   <- sample(departments, 1)
      perf   <- round(max(0, min(100, rnorm(1, 70, 15))), 1)
      kpi    <- if (perf > 75) round(runif(1, 91, 100), 1) else round(runif(1, 50, 89), 1)
      rows_list[[idx]] <- data.frame(company_id = c, employee_id = e, salary = salary,
                                     department = dept, performance_score = perf,
                                     KPI_score = kpi, stringsAsFactors = FALSE)
      idx <- idx + 1L
    }
  }
  do.call(rbind, rows_list)
}

company_data_t6 <- generate_company_data_t6(4, 20)

normalize_columns <- function(df) {
  df_norm <- df
  for (col in names(df_norm)) {
    if (is.numeric(df_norm[[col]])) {
      col_min <- min(df_norm[[col]], na.rm = TRUE)
      col_max <- max(df_norm[[col]], na.rm = TRUE)
      if (col_max != col_min)
        df_norm[[col]] <- round((df_norm[[col]] - col_min) / (col_max - col_min), 4)
    }
  }
  return(df_norm)
}

z_score <- function(df) {
  df_z <- df
  for (col in names(df_z)) {
    if (is.numeric(df_z[[col]])) {
      col_mean <- mean(df_z[[col]], na.rm = TRUE)
      col_sd   <- sd(df_z[[col]],   na.rm = TRUE)
      if (col_sd != 0)
        df_z[[col]] <- round((df_z[[col]] - col_mean) / col_sd, 4)
    }
  }
  return(df_z)
}

df_norm_t6 <- normalize_columns(company_data_t6)
df_z_t6    <- z_score(company_data_t6)

cat("=== Original (first 5 rows) ===\n")

## === Original (first 5 rows) ===

head(company_data_t6[, c("salary","performance_score","KPI_score")], 5)

cat("=== After min-max normalisation ===\n")

## === After min-max normalisation ===

head(df_norm_t6[, c("salary","performance_score","KPI_score")], 5)

cat("=== After Z-score standardisation ===\n")

## === After Z-score standardisation ===

head(df_z_t6[, c("salary","performance_score","KPI_score")], 5)

---

Task 6.2 — Feature Engineering

— R

company_data_t6$performance_category <- NA_character_
company_data_t6$salary_bracket       <- NA_character_

for (i in seq_len(nrow(company_data_t6))) {
  perf   <- company_data_t6$performance_score[i]
  salary <- company_data_t6$salary[i]

  company_data_t6$performance_category[i] <-
    if      (perf > 80) "Excellent"
    else if (perf > 65) "Very Good"
    else if (perf > 50) "Good"
    else if (perf > 35) "Average"
    else                "Poor"

  company_data_t6$salary_bracket[i] <-
    if      (salary > 10) "High"
    else if (salary >  6) "Mid"
    else                  "Low"
}

head(company_data_t6[, c("salary","performance_score",
                           "performance_category","salary_bracket")], 8)

---

Task 6.3 — Interactive Distribution Comparison

— R

sal_compare <- data.frame(
  Original   = company_data_t6$salary,
  Normalised = df_norm_t6$salary,
  Z_Score    = df_z_t6$salary
) %>% pivot_longer(everything(), names_to = "Transformation", values_to = "Value")

perf_compare <- data.frame(
  Original   = company_data_t6$performance_score,
  Normalised = df_norm_t6$performance_score,
  Z_Score    = df_z_t6$performance_score
) %>% pivot_longer(everything(), names_to = "Transformation", values_to = "Value")

tr_colours <- c("Original"="#3498db","Normalised"="#2ecc71","Z_Score"="#e74c3c")

p_sal_hist <- plot_ly()
for (tr in c("Original","Normalised","Z_Score")) {
  vals <- sal_compare$Value[sal_compare$Transformation == tr]
  p_sal_hist <- add_trace(p_sal_hist,
    x = vals, type = "histogram", nbinsx = 15, name = tr,
    marker = list(color = tr_colours[[tr]], opacity = 0.7))
}
p_sal_hist <- layout(p_sal_hist,
  barmode = "overlay",
  title   = "Task 6: Salary Distribution — Before vs. After Transformation",
  xaxis   = list(title = "Value"), yaxis = list(title = "Count"),
  plot_bgcolor = "white", paper_bgcolor = "white")

p_perf_hist <- plot_ly()
for (tr in c("Original","Normalised","Z_Score")) {
  vals <- perf_compare$Value[perf_compare$Transformation == tr]
  p_perf_hist <- add_trace(p_perf_hist,
    x = vals, type = "histogram", nbinsx = 15, name = tr,
    marker = list(color = tr_colours[[tr]], opacity = 0.7))
}
p_perf_hist <- layout(p_perf_hist,
  barmode = "overlay",
  title   = "Task 6: Performance Score Distribution — Before vs. After Transformation",
  xaxis   = list(title = "Value"), yaxis = list(title = "Count"),
  plot_bgcolor = "white", paper_bgcolor = "white")

p_sal_box <- plot_ly(sal_compare,
  x = ~Transformation, y = ~Value,
  color = ~Transformation, colors = tr_colours,
  type  = "box"
) %>% layout(
  title  = "Task 6: Salary — Boxplot Comparison",
  xaxis  = list(title = ""), yaxis = list(title = "Value"),
  showlegend   = FALSE,
  plot_bgcolor  = "white",
  paper_bgcolor = "white")

p_sal_hist

p_perf_hist

p_sal_box

---

Task 7 — Mini Project: Company KPI Dashboard

Task 7 is the capstone mini project integrating all concepts from Tasks 1–6. The objectives are:

1. Generate a large-scale dataset for 7 companies, each with 100 employees.
2. Categorise every employee into KPI tiers using a loop.
3. Summarise per-company statistics.
4. Produce a full interactive dashboard of four charts.

Dataset Scale

Parameter	Value / Definition
Companies	7
Employees per company	100
Total records	700
KPI Tier — High	KPI_score > 85
KPI Tier — Mid	KPI_score 65–85
KPI Tier — Low	KPI_score < 65

---

Task 7.1 — Large-Scale Dataset Generation

— R

set.seed(777)

# FIX: gunakan list + do.call(rbind) untuk dataset 700 baris
generate_company_data_t7 <- function(n_company, n_employees) {
  departments <- c("HR","IT","Finance","Marketing","Operations")
  rows_list   <- vector("list", n_company * n_employees)
  idx <- 1L
  for (c in 1:n_company) {
    for (e in 1:n_employees) {
      salary <- round(runif(1, 3, 15), 2)
      dept   <- sample(departments, 1)
      perf   <- round(max(0, min(100, rnorm(1, 70, 15))), 1)
      kpi    <- if (perf > 75) round(runif(1, 85, 100), 1) else round(runif(1, 40, 84), 1)
      rows_list[[idx]] <- data.frame(company_id = c, employee_id = e, salary = salary,
                                     department = dept, performance_score = perf,
                                     KPI_score = kpi, stringsAsFactors = FALSE)
      idx <- idx + 1L
    }
  }
  do.call(rbind, rows_list)
}

kpi_df <- generate_company_data_t7(n_company = 7, n_employees = 100)
cat("Dataset:", nrow(kpi_df), "rows ×", ncol(kpi_df), "columns\n\n")

## Dataset: 700 rows × 6 columns

head(kpi_df, 6)

---

Task 7.2 — KPI Tier Assignment & Company Summary

— R

# FIX: gunakan ifelse() vectorised — jauh lebih cepat dari loop per baris untuk 700 baris
kpi_df$KPI_tier <- ifelse(kpi_df$KPI_score > 85, "High",
                    ifelse(kpi_df$KPI_score >= 65, "Mid", "Low"))

company_summary_t7 <- kpi_df %>%
  group_by(company_id) %>%
  summarise(
    Avg_Salary = round(mean(salary), 2),
    Avg_KPI    = round(mean(KPI_score), 1),
    Top_N      = sum(KPI_tier == "High"),
    Mid_N      = sum(KPI_tier == "Mid"),
    Low_N      = sum(KPI_tier == "Low"),
    .groups = "drop"
  )

kable(company_summary_t7,
      col.names = c("Company","Avg Salary (k)","Avg KPI","KPI High","KPI Mid","KPI Low"),
      align     = "crrrrrr",
      caption   = "KPI dashboard summary: 7 companies × 100 employees each")

KPI dashboard summary: 7 companies × 100 employees each

Company	Avg Salary (k)	Avg KPI	KPI High	KPI Mid	KPI Low
1	8.80	70.9	30	30	40
2	8.99	73.7	33	34	33
3	8.34	73.5	37	26	37
4	8.30	73.9	39	26	35
5	8.88	73.7	36	30	34
6	8.58	74.9	45	24	31
7	8.43	70.5	32	24	44

---

Task 7.3 — Interactive KPI Dashboard

— R

tier_colours <- c("High"="#2ecc71","Mid"="#f39c12","Low"="#e74c3c")

tier_counts <- kpi_df %>%
  group_by(company_id, KPI_tier) %>%
  summarise(count = n(), .groups = "drop")

p_tier <- plot_ly(tier_counts,
  x      = ~factor(company_id), y = ~count,
  color  = ~KPI_tier, colors = tier_colours,
  type   = "bar",
  hovertemplate = "Company %{x} | Tier: %{color} | Count: %{y}<extra></extra>"
) %>% layout(
  barmode = "group",
  title   = "Chart 1: KPI Tier Distribution per Company",
  xaxis   = list(title = "Company ID"),
  yaxis   = list(title = "Number of Employees"),
  legend  = list(title = list(text = "<b>KPI Tier</b>")),
  plot_bgcolor  = "white",
  paper_bgcolor = "white")

dept_top <- kpi_df %>%
  filter(KPI_tier == "High") %>%
  group_by(department) %>%
  summarise(top_count = n(), .groups = "drop") %>%
  arrange(top_count)

p_dept <- plot_ly(dept_top,
  y = ~department, x = ~top_count,
  type = "bar", orientation = "h",
  marker = list(color = "#3498db", opacity = 0.85),
  hovertemplate = "%{y}: %{x} top performers<extra></extra>"
) %>% layout(
  title  = "Chart 2: Top Performers by Department (KPI > 85)",
  xaxis  = list(title = "Count of Top Performers"),
  yaxis  = list(title = ""),
  plot_bgcolor  = "white",
  paper_bgcolor = "white")

kpi_df$company_label <- paste("Company", kpi_df$company_id)

p_salary <- plot_ly(kpi_df,
  x = ~company_label, y = ~salary,
  color = ~company_label, type = "box", boxpoints = "outliers",
  hovertemplate = "%{x}<br>Salary: %{y:.2f}k<extra></extra>"
) %>% layout(
  title      = "Chart 3: Salary Distribution per Company",
  xaxis      = list(title = "Company"),
  yaxis      = list(title = "Salary (thousands)"),
  showlegend = FALSE,
  plot_bgcolor  = "white",
  paper_bgcolor = "white")

p_perf_kpi <- plot_ly(kpi_df,
  x = ~performance_score, y = ~KPI_score,
  color = ~company_label, type = "scatter", mode = "markers",
  marker = list(size = 5, opacity = 0.6),
  hovertemplate = "<b>%{color}</b><br>Performance: %{x}<br>KPI: %{y}<extra></extra>"
) %>% layout(
  title     = "Chart 4: Performance Score vs. KPI Score",
  xaxis     = list(title = "Performance Score"),
  yaxis     = list(title = "KPI Score"),
  legend    = list(title = list(text = "<b>Company</b>")),
  hovermode = "closest",
  plot_bgcolor  = "white",
  paper_bgcolor = "white")

p_tier

p_dept

p_salary

p_perf_kpi

---

Task 8 (Bonus) — Automated Report Generation

The objective of Task 8 is to demonstrate automated report generation: one function encapsulates all reporting logic for a single company, and a loop calls it for every company — producing consistent, structured outputs at scale without any repetitive code.

---

Task 8.1 — Report Generation Function

— R

set.seed(777)

# FIX: gunakan list + do.call(rbind) untuk dataset 700 baris
generate_company_data_t8 <- function(n_company, n_employees) {
  departments <- c("HR","IT","Finance","Marketing","Operations")
  rows_list   <- vector("list", n_company * n_employees)
  idx <- 1L
  for (c in 1:n_company) {
    for (e in 1:n_employees) {
      salary <- round(runif(1, 3, 15), 2)
      dept   <- sample(departments, 1)
      perf   <- round(max(0, min(100, rnorm(1, 70, 15))), 1)
      kpi    <- if (perf > 75) round(runif(1, 85, 100), 1) else round(runif(1, 40, 84), 1)
      rows_list[[idx]] <- data.frame(company_id = c, employee_id = e, salary = salary,
                                     department = dept, performance_score = perf,
                                     KPI_score = kpi, stringsAsFactors = FALSE)
      idx <- idx + 1L
    }
  }
  do.call(rbind, rows_list)
}

kpi_df_t8 <- generate_company_data_t8(7, 100)
kpi_df_t8$KPI_tier <- ifelse(kpi_df_t8$KPI_score > 85, "High",
                       ifelse(kpi_df_t8$KPI_score >= 65, "Mid", "Low"))

generate_company_report <- function(df, cid) {
  co_df        <- df %>% filter(company_id == cid)
  tier_colours <- c("High"="#2ecc71","Mid"="#f39c12","Low"="#e74c3c")

  stats <- data.frame(
    Metric = c("Total Employees","Avg Salary (k)","Avg Performance Score",
               "Avg KPI Score","KPI High (>85)","KPI Mid (65–85)","KPI Low (<65)"),
    Value  = c(nrow(co_df),
               round(mean(co_df$salary), 2),
               round(mean(co_df$performance_score), 1),
               round(mean(co_df$KPI_score), 1),
               sum(co_df$KPI_tier == "High"),
               sum(co_df$KPI_tier == "Mid"),
               sum(co_df$KPI_tier == "Low")))

  cat("\n══════════════════════════════════════════════\n")
  cat(paste0("  COMPANY ", cid, "  —  AUTOMATED KPI REPORT\n"))
  cat("══════════════════════════════════════════════\n\n")

  cat("[ Summary Statistics ]\n")
  print(kable(stats, col.names = c("Metric","Value"), align = "lr"))

  top_df <- co_df %>%
    filter(KPI_tier == "High") %>%
    select(employee_id, department, salary, performance_score, KPI_score) %>%
    arrange(desc(KPI_score)) %>%
    head(5)

  cat("\n[ Top 5 Performers ]\n")
  print(kable(top_df,
    col.names = c("Employee ID","Department","Salary (k)","Performance","KPI"),
    align     = "clrrr"))

  dept_summary <- co_df %>%
    group_by(department) %>%
    summarise(Count = n(), Avg_KPI = round(mean(KPI_score), 1), .groups = "drop")

  cat("\n[ Department Breakdown ]\n")
  print(kable(dept_summary,
    col.names = c("Department","Headcount","Avg KPI"), align = "lrr"))

  p1 <- plot_ly(co_df, x = ~department, color = ~KPI_tier, colors = tier_colours,
    type = "histogram", barnorm = "fraction",
    hovertemplate = "%{x} — %{color}: %{y:.1%}<extra></extra>"
  ) %>% layout(
    title   = paste0("Company ", cid, ": KPI Tier Proportion by Department"),
    xaxis   = list(title = "Department"),
    yaxis   = list(title = "Proportion", tickformat = ".0%"),
    barmode = "stack",
    plot_bgcolor  = "white",
    paper_bgcolor = "white")

  p2 <- plot_ly(co_df, x = ~salary, color = ~KPI_tier, colors = tier_colours,
    type = "histogram", nbinsx = 15, opacity = 0.8,
    hovertemplate = "Salary: %{x:.1f}k<br>Count: %{y}<extra></extra>"
  ) %>% layout(
    barmode = "overlay",
    title   = paste0("Company ", cid, ": Salary Distribution by KPI Tier"),
    xaxis   = list(title = "Salary (thousands)"),
    yaxis   = list(title = "Count"),
    plot_bgcolor  = "white",
    paper_bgcolor = "white")

  print(p1)
  print(p2)

  # FIX: CSV hanya ditulis jika belum ada — mencegah duplikasi di setiap run/knit
  folder    <- "reports"
  if (!dir.exists(folder)) dir.create(folder)
  file_name <- file.path(folder, paste0("company_", cid, "_report.csv"))

  if (!file.exists(file_name)) {
    write.csv(co_df, file = file_name, row.names = FALSE)
    cat(paste0("\n  CSV exported: ", file_name, "\n\n"))
  } else {
    cat(paste0("\n  CSV sudah ada, skip: ", file_name, "\n\n"))
  }
}

---

Task 8.2 — Automated Loop: One Loop, Seven Reports

— R

# A single for loop produces a complete report for every company automatically.
for (cid in sort(unique(kpi_df_t8$company_id))) {
  generate_company_report(kpi_df_t8, cid)
}

## 
## ══════════════════════════════════════════════
##   COMPANY 1  —  AUTOMATED KPI REPORT
## ══════════════════════════════════════════════
## 
## [ Summary Statistics ]
## 
## 
## |Metric                | Value|
## |:---------------------|-----:|
## |Total Employees       | 100.0|
## |Avg Salary (k)        |   8.8|
## |Avg Performance Score |  69.7|
## |Avg KPI Score         |  70.9|
## |KPI High (>85)        |  30.0|
## |KPI Mid (65–85)       |  30.0|
## |KPI Low (<65)         |  40.0|
## 
## [ Top 5 Performers ]
## 
## 
## | Employee ID |Department | Salary (k)| Performance|  KPI|
## |:-----------:|:----------|----------:|-----------:|----:|
## |     49      |Operations |       5.26|        94.8| 99.9|
## |     94      |Marketing  |       3.00|        89.1| 99.7|
## |     86      |Finance    |       9.80|        85.6| 99.0|
## |      8      |HR         |       6.98|       100.0| 98.6|
## |     79      |Marketing  |       3.61|        79.6| 98.4|
## 
## [ Department Breakdown ]
## 
## 
## |Department | Headcount| Avg KPI|
## |:----------|---------:|-------:|
## |Finance    |        22|    61.2|
## |HR         |        19|    77.1|
## |IT         |        25|    70.9|
## |Marketing  |        18|    73.9|
## |Operations |        16|    73.7|

## 
##   CSV sudah ada, skip: reports/company_1_report.csv
## 
## 
## ══════════════════════════════════════════════
##   COMPANY 2  —  AUTOMATED KPI REPORT
## ══════════════════════════════════════════════
## 
## [ Summary Statistics ]
## 
## 
## |Metric                |  Value|
## |:---------------------|------:|
## |Total Employees       | 100.00|
## |Avg Salary (k)        |   8.99|
## |Avg Performance Score |  70.40|
## |Avg KPI Score         |  73.70|
## |KPI High (>85)        |  33.00|
## |KPI Mid (65–85)       |  34.00|
## |KPI Low (<65)         |  33.00|
## 
## [ Top 5 Performers ]
## 
## 
## | Employee ID |Department | Salary (k)| Performance|  KPI|
## |:-----------:|:----------|----------:|-----------:|----:|
## |     85      |Marketing  |      12.87|        82.0| 99.8|
## |     93      |Marketing  |      11.43|        88.2| 99.0|
## |     97      |Operations |      12.72|        81.6| 98.9|
## |     98      |Marketing  |       4.03|        92.7| 98.8|
## |     39      |Marketing  |       8.39|        83.5| 98.2|
## 
## [ Department Breakdown ]
## 
## 
## |Department | Headcount| Avg KPI|
## |:----------|---------:|-------:|
## |Finance    |        24|    72.5|
## |HR         |        17|    72.4|
## |IT         |        22|    69.8|
## |Marketing  |        22|    77.9|
## |Operations |        15|    76.5|

## 
##   CSV sudah ada, skip: reports/company_2_report.csv
## 
## 
## ══════════════════════════════════════════════
##   COMPANY 3  —  AUTOMATED KPI REPORT
## ══════════════════════════════════════════════
## 
## [ Summary Statistics ]
## 
## 
## |Metric                |  Value|
## |:---------------------|------:|
## |Total Employees       | 100.00|
## |Avg Salary (k)        |   8.34|
## |Avg Performance Score |  70.50|
## |Avg KPI Score         |  73.50|
## |KPI High (>85)        |  37.00|
## |KPI Mid (65–85)       |  26.00|
## |KPI Low (<65)         |  37.00|
## 
## [ Top 5 Performers ]
## 
## 
## | Employee ID |Department | Salary (k)| Performance|  KPI|
## |:-----------:|:----------|----------:|-----------:|----:|
## |     51      |Marketing  |       3.84|       100.0| 99.7|
## |     70      |Operations |      13.13|        79.8| 99.7|
## |     11      |Operations |       6.45|        84.5| 99.3|
## |     59      |Finance    |       5.62|        90.1| 98.9|
## |     26      |HR         |      13.20|        79.9| 98.7|
## 
## [ Department Breakdown ]
## 
## 
## |Department | Headcount| Avg KPI|
## |:----------|---------:|-------:|
## |Finance    |        23|    69.7|
## |HR         |        17|    81.0|
## |IT         |        20|    70.2|
## |Marketing  |        21|    73.8|
## |Operations |        19|    74.6|

## 
##   CSV sudah ada, skip: reports/company_3_report.csv
## 
## 
## ══════════════════════════════════════════════
##   COMPANY 4  —  AUTOMATED KPI REPORT
## ══════════════════════════════════════════════
## 
## [ Summary Statistics ]
## 
## 
## |Metric                | Value|
## |:---------------------|-----:|
## |Total Employees       | 100.0|
## |Avg Salary (k)        |   8.3|
## |Avg Performance Score |  70.4|
## |Avg KPI Score         |  73.9|
## |KPI High (>85)        |  39.0|
## |KPI Mid (65–85)       |  26.0|
## |KPI Low (<65)         |  35.0|
## 
## [ Top 5 Performers ]
## 
## 
## | Employee ID |Department | Salary (k)| Performance|  KPI|
## |:-----------:|:----------|----------:|-----------:|----:|
## |     37      |Operations |      13.87|        75.9| 99.4|
## |     19      |HR         |       6.70|        75.6| 99.0|
## |     79      |Operations |      14.66|       100.0| 98.6|
## |     72      |Marketing  |      14.38|        77.0| 98.0|
## |     28      |Finance    |       3.52|        90.2| 97.2|
## 
## [ Department Breakdown ]
## 
## 
## |Department | Headcount| Avg KPI|
## |:----------|---------:|-------:|
## |Finance    |        19|    75.2|
## |HR         |        24|    72.6|
## |IT         |        16|    73.5|
## |Marketing  |        15|    77.4|
## |Operations |        26|    72.5|

## 
##   CSV sudah ada, skip: reports/company_4_report.csv
## 
## 
## ══════════════════════════════════════════════
##   COMPANY 5  —  AUTOMATED KPI REPORT
## ══════════════════════════════════════════════
## 
## [ Summary Statistics ]
## 
## 
## |Metric                |  Value|
## |:---------------------|------:|
## |Total Employees       | 100.00|
## |Avg Salary (k)        |   8.88|
## |Avg Performance Score |  70.10|
## |Avg KPI Score         |  73.70|
## |KPI High (>85)        |  36.00|
## |KPI Mid (65–85)       |  30.00|
## |KPI Low (<65)         |  34.00|
## 
## [ Top 5 Performers ]
## 
## 
## | Employee ID |Department | Salary (k)| Performance|   KPI|
## |:-----------:|:----------|----------:|-----------:|-----:|
## |     47      |Marketing  |       8.21|        79.2| 100.0|
## |      6      |Finance    |      10.40|        86.7|  99.8|
## |     67      |IT         |      11.92|        85.5|  99.7|
## |     98      |HR         |       7.73|        76.9|  99.5|
## |     86      |IT         |      13.85|        76.8|  99.1|
## 
## [ Department Breakdown ]
## 
## 
## |Department | Headcount| Avg KPI|
## |:----------|---------:|-------:|
## |Finance    |        21|    74.9|
## |HR         |        27|    73.3|
## |IT         |        23|    72.8|
## |Marketing  |        16|    74.9|
## |Operations |        13|    73.1|

## 
##   CSV sudah ada, skip: reports/company_5_report.csv
## 
## 
## ══════════════════════════════════════════════
##   COMPANY 6  —  AUTOMATED KPI REPORT
## ══════════════════════════════════════════════
## 
## [ Summary Statistics ]
## 
## 
## |Metric                |  Value|
## |:---------------------|------:|
## |Total Employees       | 100.00|
## |Avg Salary (k)        |   8.58|
## |Avg Performance Score |  71.20|
## |Avg KPI Score         |  74.90|
## |KPI High (>85)        |  45.00|
## |KPI Mid (65–85)       |  24.00|
## |KPI Low (<65)         |  31.00|
## 
## [ Top 5 Performers ]
## 
## 
## | Employee ID |Department | Salary (k)| Performance|  KPI|
## |:-----------:|:----------|----------:|-----------:|----:|
## |     51      |IT         |      14.91|        77.1| 99.5|
## |     19      |Finance    |      11.75|        80.5| 99.1|
## |     25      |Operations |       5.88|        78.2| 99.0|
## |     34      |IT         |       8.45|        78.5| 98.3|
## |     41      |IT         |       7.79|       100.0| 98.1|
## 
## [ Department Breakdown ]
## 
## 
## |Department | Headcount| Avg KPI|
## |:----------|---------:|-------:|
## |Finance    |        18|    75.7|
## |HR         |        22|    71.2|
## |IT         |        21|    80.3|
## |Marketing  |        20|    75.7|
## |Operations |        19|    71.6|

## 
##   CSV sudah ada, skip: reports/company_6_report.csv
## 
## 
## ══════════════════════════════════════════════
##   COMPANY 7  —  AUTOMATED KPI REPORT
## ══════════════════════════════════════════════
## 
## [ Summary Statistics ]
## 
## 
## |Metric                |  Value|
## |:---------------------|------:|
## |Total Employees       | 100.00|
## |Avg Salary (k)        |   8.43|
## |Avg Performance Score |  69.40|
## |Avg KPI Score         |  70.50|
## |KPI High (>85)        |  32.00|
## |KPI Mid (65–85)       |  24.00|
## |KPI Low (<65)         |  44.00|
## 
## [ Top 5 Performers ]
## 
## 
## | Employee ID |Department | Salary (k)| Performance|  KPI|
## |:-----------:|:----------|----------:|-----------:|----:|
## |     79      |Operations |       3.18|        77.7| 99.0|
## |     11      |Operations |       9.97|        90.2| 98.9|
## |     64      |Marketing  |       5.36|        77.8| 98.0|
## |     40      |HR         |      12.21|        95.0| 97.9|
## |     86      |Finance    |      13.85|        76.2| 97.9|
## 
## [ Department Breakdown ]
## 
## 
## |Department | Headcount| Avg KPI|
## |:----------|---------:|-------:|
## |Finance    |        26|    69.9|
## |HR         |        22|    71.1|
## |IT         |        13|    72.2|
## |Marketing  |        19|    74.2|
## |Operations |        20|    66.0|

## 
##   CSV sudah ada, skip: reports/company_7_report.csv

---

Conclusion

This practicum has demonstrated, in a progressive and integrated manner, how functions, loops, and conditional branching are sufficient to build a complete, end-to-end data science pipeline.

Tasks 1–2 established the foundation: well-validated functions and nested loops for formula evaluation and sales simulation. Tasks 3–4 extended the pattern to classification and hierarchical data generation, reflecting real-world business datasets. Task 5 applied the same loop-and-condition skeleton to stochastic simulation, proving that Monte Carlo methods require no specialised libraries — only random sampling and counting. Task 6 addressed the pre-modelling pipeline through data transformation and feature engineering, confirming that linear transformations preserve distributional shape while rescaling the axis. Task 7 consolidated all prior concepts into a full KPI dashboard at scale (700 records, 7 companies), demonstrating that the same code patterns scale effortlessly. Task 8 completed the cycle with automated report generation, where a single function called by a loop produces consistent, structured outputs for every entity — the hallmark of production-ready reporting pipelines.

Throughout all tasks, R and Python implementations were maintained in parallel. While syntax differs — else if vs. elif, rbind() vs. list.append(), base R vs. pandas — the underlying logic is identical, reinforcing language-agnostic programming thinking that is essential in modern data science practice.

---

References

1. Functions and Loops — https://bookdown.org/dsciencelabs/data_science_programming/03-Functions-and-Loops.html
2. Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York.
3. Sievert, C. (2020). Interactive Web-Based Data Visualisation with R, plotly, and shiny. CRC Press.
4. Plotly Technologies Inc. (2015). Collaborative data science. https://plot.ly
5. R Core Team (2023). R: A Language and Environment for Statistical Computing. R Foundation.
6. Van Rossum, G., & Drake, F. L. (2009). Python 3 Reference Manual. CreateSpace.