COMP3260Data Science Foundations

Module 4 · 3 units · 25 tutorials · Weeks 9–11

Statistical Foundations for Data Science

Probability, sampling, inference and the reasoning behind statistical claims. How to move from describing what you observed to reasoning about what it means.

Start Unit 10 All modules

Module overview

What this module covers

Module 4 introduces the statistical reasoning that separates descriptive analysis from inference. You learn how probability models uncertainty, how sampling allows conclusions about populations, and how hypothesis testing provides a framework for evidence.

The module closes with regression treated as a statistical technique for describing relationships — not as a machine learning method. Predictive modelling is introduced in COMP328.

Statistical regression only

This is statistical regression for inference and description. Machine learning regression is deliberately deferred to COMP328.

At a glance

Units
3 (U10–U12)
Lessons
25
Weeks
9–11
Assessment
3 unit quizzes · Assignment 3 due Week 11 · Assignment 4 released Week 11
Key tools
NumPy, SciPy, statsmodels, pandas

Learning outcomes

By the end of Module 4 you will be able to

Reason probabilistically

  • Apply the rules of probability.
  • Work with common distributions.
  • Update beliefs using conditional probability and Bayes' theorem.

Infer

  • Explain sampling variability and the central limit theorem.
  • Construct and interpret confidence intervals.
  • Conduct hypothesis tests and interpret p-values correctly.

Model relationships

  • Distinguish correlation from causation.
  • Fit and interpret simple and multiple linear regression.
  • Evaluate and communicate model assumptions.

Units

Three units in this module

👤