Forecasting Newsletter: July 2020.

post by NunoSempere · 2020-08-01T16:56:41.600Z · EA · GW · 1 comments

Contents

  Highlights
  Index
  Prediction Markets & Forecasting Platforms.
    undertakings
  Negative Examples.
  News & Hard to Categorize Content.
  Long Content.
None
1 comment

Highlights

Index

Sign up here or browse past newsletters here.

Prediction Markets & Forecasting Platforms.

Ordered in subjective order of importance:

	# Variable: Number of ice creams an unsupervised child has consumed, 
	# when left alone in an ice cream shop.   

	# Current time (hours passed)  
	t=10  

	# Scenario with lots of uncertainty
	w_1 = 0.75 ## Weight for this scenario.
	min_uncertain(t) = t*2
	max_uncertain(t) = t*20

	# Optimistic scenario
	w_2 = 0.25 ## Weight for the optimistic scenario
	min_optimistic(t) = 1*t
	max_optimistic(t) = 3*t
	mean(t) = (min_optimistic(t) + max_optimistic(t)/2)
	stdev(t) = t*(2)^(1/2)

	# Overall guess
	## A long-tailed lognormal for the uncertain scenario
	## and a tight normal for the optimistic scenario

	mm(min_uncertain(t) to max_uncertain(t), normal(mean(t), stdev(t)), [w_1, w_2])

	## Compare with: mm(2 to 20, normal(2, 1.4142), [0.75, 0.25])

New undertakings

Negative Examples.

News & Hard to Categorize Content.

Long Content.


Note to the future: All links are added automatically to the Internet Archive. In case of link rot, go here


"horses for courses, the way you do the forecast, the way you present it depends on what you're trying to achieve with it"


1 comments

Comments sorted by top scores.

comment by meerpirat · 2020-08-14T09:30:22.021Z · EA(p) · GW(p)

Thanks, I still find this very useful. Haven't seen Ought's Ergo before, really cool.