# Questions to consider

- Is impact heavy-tailed across interventions? To what extent?
- Is impact heavy-tailed across fields? To what extent?
- Is impact heavy-tailed across causes? To what extent?
*For mathematicians: how might we characterise heavy-tailedness?*

# Resources

As this is an optional week, the reading gives a looser summary. You're encouraged to go research other information on this topic! Start with the articles that most interest you, and read as much as you can from there.

- The Zipf Mystery: Fun VSauce video on Zipf Law and Pareto Principle (VSauce, 2015) (video - 21 mins).
- Heavy Tails and Altruism: When Your Intuition Fails You: Short explainer on heavy tails and relation to altruism (Jan Michelfeit, 2020) (post - 11 mins).
- Why Charities Usually Don't Differ Astronomically in Expected Cost-Effectiveness: A similar explainer on heavy tails (Brian Tomasik, 2017) (post - 40 mins).
- Is most expected suffering due to worst-case outcomes? Is suffering heavy-tailed? (S-risks.org, 2021) (report - 8 mins).
- Catastrophes, Conspiracies, and Subexponential Distributions (Part II): A possible mathematical definition for heavy-tailed distributions (Adam Wierman, 2013) (post - 4 mins).
- Benford’s law, Zipf’s law, and the Pareto distribution: Why heavy tails might be common (Terence Tao, 2009) (post - 19 mins).
- On characterising heavy-tailedness: “There are many formalizations of heavy-tailedness out there. I define five intuititive principles that I expect a good definition to satisfy: action-relevance, distinguish negative and positive risk, allow finite support, apply to empirically observed phenomena and provide a characterization in terms of a universal class of distributions. I discuss each in turn and provide examples.” (Jaime Sevilla Molina, 2020) (post - 6 mins).