Statistics¶
A book in 20 chapters about learning from data: from reading a graph honestly to confidence intervals, regression, Bayesian methods, causal inference and the replication crisis. Each chapter starts from a question, proves what it claims, works examples step by step, and ends with questions whose answers are folded underneath.
How to read it¶
- In order. Each chapter needs only the ones before it; the Next link at the end of every chapter walks you through.
- Levels inside a chapter. The box at the top of each chapter says which sections are beginner, intermediate, advanced or expert. On a first read, skip the harder sections and come back later.
- Probability behind it. Parts I and II need only school algebra. From Part II on, the Probability book supplies the distributions, the law of large numbers and the central limit theorem; from Part III on, calculus appears.
- Slower help. The first chapters point to the matching lessons on the Topics shelf, which teach the same ideas more gently, for younger readers.
- Honest data. Every data set is either a well-known real one, described as it was published, or clearly marked as made up for the example.
Contents¶
| Part | Chapter | Level |
|---|---|---|
| I. Describing data | 1. What statistics is for | beginner |
| 2. Pictures of data | beginner | |
| 3. Summaries | beginner | |
| 4. Relationships and correlation | beginner to intermediate | |
| II. Collecting data and sampling | 5. Sampling and experiments | intermediate |
| 6. Sampling distributions | intermediate | |
| 7. Estimation | intermediate | |
| 8. Confidence intervals | intermediate | |
| III. Testing and models | 9. Hypothesis tests | intermediate |
| 10. Categorical data | intermediate | |
| 11. Linear regression | intermediate to advanced | |
| 12. Multiple regression | advanced | |
| IV. Deeper inference | 13. Likelihood theory | advanced |
| 14. Bayesian statistics | advanced | |
| 15. Nonparametric methods | advanced | |
| 16. Time series | advanced | |
| V. Modern statistics | 17. Experimental design and causal inference | advanced to expert |
| 18. High-dimensional statistics | expert | |
| 19. Statistical learning | advanced to expert | |
| 20. Where statistics goes | expert |
The glossary defines every term, with a link to the chapter that teaches it.
Three ways in¶
- To read the news and studies with a clear head: chapters 1 to 5, then 9.
- To analyse data yourself: chapters 6 to 12.
- For the theory and today's methods: chapters 13 to 20, with the Probability book beside you.