This repository contains code written for my CS109 challenge project, "Check for Your Privilege." There are three components:
- A library for calculating probability using a Bayesian network with rejection sampling
- Base class is in model/include/model.h
- Subclasses are used to define the probabilities for individual models, such as model/include/SeattleModel.h
- I ended up not using this because I decided to focus on conditional probabilities, but thought I'd include the code anyway
- Jupyter notebook for processing Public Use Microdata Sample data from the 2019 American Community Survey
- Reads in data and adds flags for factors of interest
- Runs logistic regressions on data
- Summarizes group counts
- Exports group counts data to CSV for use in Observable via SQLite
- Observable code:
/observable- Back the notebook with the interactive tool: https://observablehq.com/@chaya/check-for-your-privilege