Data Science for Economists

Hands-on learning with cutting-edge tools and methodologies, from web scraping to satellite imagery analysis.

Author

Julian Hinz

Fields in Kazakhstan.

About

This short course introduces economists to modern data science tools and workflows. Over two intensive days, participants learn to work with large structured datasets, scrape the web for novel data, analyze text and spatial information, and integrate large language models into their research pipelines — all in R.

Each module pairs a lecture introducing research applications with a hands-on coding session where participants replicate the analysis on real data. By the end, you will have a working toolkit that extends well beyond standard econometrics.

This is the ASP edition of a longer course taught at Bielefeld University jointly with Irene Iodice (irene.iodice@uni-bielefeld.de).

Course repository: github.com/julianhinz/data-science-ASP-2026

Schedule

Morning (9:00–12:30) Afternoon (13:30–17:00)
Tue, March 3 01 Getting Started + 02 Toolkit 03 Large Structured Data + 04 Web Scraping
Wed, March 4 07 Spatial & Satellite Data + 06 Text as Data 10 LLMs + 12 AI-Assisted Research

Modules

  1. Getting Started — course overview, reproducibility, and modern AI tools
  2. Toolkit: R and the Shell — R programming, Unix shell, git, and make
  3. Large Structured Data — data.table, DuckDB, and Apache Arrow
  4. Web Scraping & APIs — rvest, APIs, and the Billion Prices Project
  5. Text as Data — tokenization, tf-idf, and sentiment analysis
  6. Spatial & Satellite Data — sf, terra, nighttime lights, and deforestation
  7. LLMs — using LLMs via APIs, structured outputs, and prompt engineering
  8. AI-Assisted Research — CLAUDE.md, skills, code review, and research life hacks

References and additional resources

Contact

Julian Hinz: julian.hinz@kielinstitut.de