The long-run average of anything random — the quantity every model and every agent is quietly optimising.
Module 5 · 0/9 done
§ 01
What happens on average
One roll of a die could be anything from 1 to 6. But roll it ten thousand times and the average settles down to something very predictable. That long-run average is the expected value. Work it out by weighting each outcome by its probability:
E[X]=i∑pixi
Fig 5.7Play a game over and over. The running average wobbles, then locks onto the expected value (dashed). This is the law of large numbers.