R for Data Science

Context: FIT1043_MOC · the unit’s second language for the value chain · the statistician’s counterpart to Python

Quick Revision

• 🎯 Objective: justify R as a data-analysis language and set it up ➔ statistics-first, interpreted, open-source.
• ⚡ Key Constraint: R vs Python is a fit-for-purpose choice — R for statistics/analysis, Python for building/deploying systems.

📝 Core

• What R is ➔ a language for analysing and visualising data; interpreted (scripting — no compile step); designed by statisticians; open-source and very popular.
• Style ➔ functional, with more data-analysis and statistical support built in (Python is object-oriented and relies more on packages).
• Setup ➔ RStudio Cloud (browser) or install R (r-project.org) + the RStudio IDE; run via the R console/RStudio or R in a shell.

⚖️ Core Decision Matrix

Aspect	R	Python
Objective	data analysis & statistics	deployment & production
Strength	ready statistical libraries	building models from scratch
Packages	tidyverse, ggplot2, caret, zoo	pandas, scikit-learn, TensorFlow
Weakness	slower, steep curve, dependency clashes	fewer specialised stats packages

⚠️ Common Mistakes

• 💡 Pick by task, not fashion ➔ statistics/exploratory analysis leans R; production pipelines/general programming lean Python (no single best tool).
• 💡 Interpreted ≠ fast ➔ R trades raw speed for interactive, statistics-friendly scripting.

🧠 Active Recall

Why choose R over Python, and what kind of language is it?

Answer

• Short answer: R is an interpreted, open-source, functional language designed by statisticians for data analysis/visualisation, with rich statistical support built in; prefer it for statistics-heavy work, Python for deployment/production.
• Why: Statistics-first ➔ more analysis/stats is native to R; Python leans on packages and suits general system building.