Act as an expert Python Data Scientist and Performance Engineer. Your task is to refactor and optimize the provided Python script for maximum efficiency, readability, and memory management. \n\n### Optimization Guidelines:\n1. Vectorization: Replace explicit for-loops with NumPy or Pandas vectorized operations wherever possible to leverage low-level optimizations.\n2. Memory Management: Optimize data types (e.g., converting objects to categories, downcasting numeric types) and ensure efficient handling of large DataFrames.\n3. Execution Speed: Identify bottlenecks in data processing and replace them with more performant alternatives (e.g., using .at/.iat instead of .loc/.iloc for scalar access).\n4. Code Quality: Ensure the code adheres to PEP 8 standards, improves modularity, and includes meaningful docstrings.\n5. Standard Libraries: Utilize Python standard libraries or specialized data science packages (like Bottleneck or NumExpr) to accelerate calculations.\n\n### Output Requirements:\n- Refactored Code: Provide the complete, optimized Python script.\n- Performance Summary: A concise list of the specific changes made and why they improve performance.\n- Complexity Comparison: A brief comparison of the time and memory complexity between the original and optimized versions.\n\n### Input Code:\n[INSERT YOUR PYTHON SCRIPT HERE]