3  About the project

3.1 The book

Pictures of Inference is a book in progress. It is meant as a single integrated text that takes a reader from “what is a correlation” to the working knowledge needed to read, evaluate, and contribute to research published in journals like the American Economic Review, Quarterly Journal of Economics, and Econometrica.

The pedagogical commitments are stated up front and respected throughout:

Visualization is primary. Every concept gets a figure, and the prose points at it. Captions interpret rather than label. The reader internalizes a consistent visual vocabulary by chapter five and benefits from it for the rest of the journey.

Math is taught as a language. Notation is parsed before it is computed. Translation between English and mathematics is an explicit, drilled skill. Equations are read aloud the first time they appear.

Tool fluency, not tool tribalism. R, Stata, Python, and others appear where they fit best. The reader develops a portfolio of fluencies rather than commitment to a single ecosystem.

Voice as a working narrator. The narrator is a wise, curious, occasionally amused traveling companion. The reader is presumed intelligent. Authority is never performed because the writing demonstrates competence by being competent.

3.2 The authors

Dr. Ian Helfrich is an economist working at the intersection of international economics, geospatial methods, and quantitative pedagogy. PhD, Georgia Institute of Technology. He works on trade-cost geometry, the measurement of trade frictions under sanctions and conflict, and the practical question of how to teach quantitative methods without losing the audience.

Dr. Elizaveta Gonchar is the co-author. (Bio to be filled in.)

3.3 The repository

The book’s source lives at a public GitHub repository. The repository contains:

  • Every chapter’s .qmd source, ready to re-render.
  • The Python figure generation code.
  • The placeholder trade dataset and its codebook.
  • The Quarto build configuration and CSS.
  • A continuous-integration workflow that rebuilds and deploys on every push.

The book is licensed CC-BY-SA 4.0 for prose and MIT for code and figures. Datasets are licensed individually and noted in their respective directories. Corrections and contributions are invited through pull requests.

3.4 Comparable works

Pictures of Inference sits alongside several existing texts and acknowledges its inheritances explicitly.

  • Statistical Rethinking (McElreath) for voice, integration of code, and accessibility-with-depth.
  • Mostly Harmless Econometrics and Mastering ’Metrics (Angrist & Pischke) for modern empirical sensibility.
  • R for Data Science (Wickham & Grolemund) for production quality, format, and reader experience.
  • The Visual Display of Quantitative Information and Envisioning Information (Tufte) for visual design philosophy.
  • Causal Inference: The Mixtape (Cunningham) for accessible voice on advanced material.
  • Econometric Analysis (Greene) for comprehensiveness and scholarly tone.

3.5 How to cite

If you draw on this book in your own work, the suggested citation is:

Helfrich, I., and Gonchar, E. (2026). Pictures of Inference: A Visual Course in Statistics, Calculus, and Econometrics. (Working draft.) Available at: ianhelfrich.com.

A formal citation will be added when the book reaches its first stable release.

3.6 How to contribute

Three ways:

  1. Corrections. Open an issue on the GitHub repository. Specific page references are welcome.
  2. Voice samples. If a chapter feels off, the underlying voice charter is in docs/voice/charter.md. Quote the offending sentence and propose an alternative.
  3. New material. Draft chapters or sections, especially for Parts II-IV, are welcome via pull request. Read docs/process.md first for the editorial workflow.

The project is multi-year and there is no shortage of work to do.