https://github.com/emjun/covid_tracking
Data analyses for tracking COVID 19 cases, testing, and mortality
https://github.com/emjun/covid_tracking
Last synced: about 1 year ago
JSON representation
Data analyses for tracking COVID 19 cases, testing, and mortality
- Host: GitHub
- URL: https://github.com/emjun/covid_tracking
- Owner: emjun
- Created: 2020-03-22T02:03:36.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2020-03-26T23:33:41.000Z (about 6 years ago)
- Last Synced: 2025-02-16T14:07:47.215Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 41 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Data source: https://covidtracking.com/ \
High-level Hypothesis statement: "Higher positive test rate and/or low numbers of tests would imply a faster rate of growth later in the positive cases curve."
This hypothesis can be broken down into Hypothesis A and Hypothesis B, below.
**Hypothesis A:** "Higher positive test rate implies faster rate of growth later in the positive cases curve"
- `hypo_a.ipynb`: Currently, Tea does not support modeling (working on providing this soon!),
so I tested a simpler hypothesis: Higher positive test counts imply higher
growth rate (as measured by increase in positive tests from yesterday, which is a metric reported in the data).
*To be totally honest, I'm not sure this is a totally accurate operationalization of the original hypothesis, even without modeling capacity.*
**Hypothesis B:** "Low numbers of tests would imply a faster rate of growth later in the positive cases curve"
- `hypo_b.ipynb`: Very similar to Hypothesis A, above. The main difference is that total number of tests, instead of positive test cases only, are considered.
**Hypthesis C:** Higher testing rates is positively related to higher death count.
Collaborators: Ben Zorn, Emery Berger