So far, we've focused on individual skills involved in a forecast: extrapolating trends [https://bounded-regret.ghost.io/forecasting-zeroth-and-first-order/], employing reference classes [https://bounded-regret.ghost.io/base-rates-and-reference-classes/], how to combine estimates [https:

For this post, I commissioned Misha Yagudin, a top-ranked forecaster, to provide feedback and commentary. I include selected comments as quotes throughout. For open-ended questions, it's easy to underestimate the number

Often there are multiple ways to forecast the same thing, and we'd like a way of combining the forecasts together. For instance, consider the following question: > Will Will be among

When we output a forecast, we're either explicitly or implicitly outputting a probability distribution. For example, if we forecast the AQI in Berkeley tomorrow to be "around" 30, plus

Part of lecture notes for the upcoming Stat157 [http://www.stat157.com/] class on Forecasting. Let's start by considering the following question: > What is the probability that Joe Biden is