You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hey all! I am a user of both Guesstimate and the (also open sourced) Manifold Markets forecasting platform.
As a feature idea, I think it would be really nice if the two platforms had an integration so that one could use the forecasts made in a Manifold market as the input for a metric in a Guesstimate model. This would allow estimates to be crowd-sourced and (ideally) updated according to the distribution of guesses of the prediction market, potentially making models more accurate.
The text was updated successfully, but these errors were encountered:
Hi! I'll comment on the generalized version of this idea first:
"forecasts and other similar data" is more general than Manifold; one possible source of such data could be Metaforecast (which has Manifold and other prediction markets in its database), but we'd also like to support other data sources, e.g. OWID
we might build a separate library or GraphQL endpoint for this purpose, because it would be useful not just for Guesstimate, but also for Squiggle
one complication is data updates; it would be costly to fetch values in real time, and also sometimes you don't want to use fresh data but need to cache the values as they were at the point when you built your model; this means that caches and UI for updates should be incorporated into the Guesstimate's UI
It probably makes sense to iterate in this direction instead of trying to get it right on the first attempt. But solving the full problem properly will be especially valuable, I think, and it's pretty high on my list of priorities.
Hey all! I am a user of both Guesstimate and the (also open sourced) Manifold Markets forecasting platform.
As a feature idea, I think it would be really nice if the two platforms had an integration so that one could use the forecasts made in a Manifold market as the input for a metric in a Guesstimate model. This would allow estimates to be crowd-sourced and (ideally) updated according to the distribution of guesses of the prediction market, potentially making models more accurate.
The text was updated successfully, but these errors were encountered: