Uncertainty wiki (Models for a non-stationary world) [WIP]

Quantifiable Uncertainty (Risk) vs Unquantifiable Uncertainty
Our blindspot: we ignore unquantifiable uncertainty.
Quantifying some of that quantifiable uncertainty is still useful
Is the world we want to model stationary, at all?
The 3 types of models
Is optimization an illusion in practice?
There are no models without assumptions
There are no models without assumptions
How about modelling when an existing relationship is likely break?

This is a a reminder for me that world is not as simple and stationary as us, data-informed people want them to be.


  • Where is unquantifiable uncertainty greatly ignored?
  • Where do we assume stationarity, no change, status quo, where that change can have extreme consequences?

I think it could be important in complex, non-deterministic systems, especially where we don’t have a lot of data, and where we need to measure phenomena over time.

  • Economics/finance
  • Social sciences

This started out as a collection of notes of Radical Uncertainty, but since then, I’ve added a lot of parallel/contradicting and my own thoughts.