Blog

Prediction of Nuclear Weapons Risk

Over the past year, I have been participating in a Forecasting Tournament through the University of Pennsylvania. This was put on by a group headlined by Phil Tetlock, father of much of the forecasting community and author of Superforecasters. This tournament had two interesting features that you don’t normally see in prediction markets; significant collaboration within and across teams and a focus on long-term existential risks. Normally prediction markets can only ask questions that resolve in the next 1-2 years, no one is particularly interested in a $20 dollar payoff after a decade of waiting. I will share the two…

Thermal Imagery Analysis: Aluminum Production in Tajikistan

By Keenan Viney With the cost of satellite launches falling quickly, high-quality imagery is becoming more available. I wanted to explore an area with some industrial application. This post will take a look at whether its possible to predict aluminum production with open source thermal imagery. I chose to look at the Talco plant in Tajikistan, because I have been to the factory and because Tajikistan is a very sunny country; clouds being the bane of much remote sensing. Gathering what production data was possible to find online and, where possible, checking multiple sources, here is the plant’s production: There…

Dynamics of National Debt

I think generally we lack mental models of dynamic systems. The debate around national debt often falls back on inappropriately applied heuristics; a government unlike a household is infinitely lived, it should be minimizing the long run tax burden and faces a borrowing constraint that is not time bound. Here I present a simple model for thinking about the national debt, it helps reason about two questions: Can a government run a persistent budget deficit? Can national debt grow forever without bankrupting the nation? Let us define some variables: D(t) is the debt at time t, with the initial debt…

An Economist’s View on Epidemiology Models

Epidemiology models need a middle way, between SEIR (with no forward-looking expectations) and agent-based simulation (too many free parameters). SEIR’s ultimately need to pin down a value for R0, but it is never a fixed value, especially in a global pandemic. The R0 value is only applicable at the very beginning of a pandemic where the parameters of the virus are basically unknown. Even in very small outbreaks there is learning that happens, after SARS-1, hospitals in effected nations established at least rudimentary protocols – this forever alters R0. This fact is already acknowledged by epidemiologists; it is rarely possible…