MENTAL MODEL · M12821

Prediction Interval

DecideHigh supportDecision Science

Included

Version 1.0.0 · Updated 2026-07-30

CORE DEFINITION

Provides the range within which future observations are likely to fall and the associated probability, as opposed to the confidence interval for estimating past parameters.

SCAFFOLDING EFFECT

psychology

Reduce cognitive load

Replace point estimates with intervals. Acknowledge that predictions inevitably have errors, and provide "most likely within this range" which is more honest and practical than "it is exactly this number."

anchor

Anchor fast decisions

Beyond the point prediction, it gives the interval into which a single future observation will fall with probability 1-α; the interval width increases with residual variance and distance from the prediction point; it quantifies the uncertainty of new observations, as opposed to the confidence interval for parameters.

MINIMUM ACTION

In progress 0/5

Practice this model in one real situation:

Fit the model to obtain predictions
calculate residual variance and prediction point uncertainty
provide the interval [ŷ ± t·se] based on the t-distribution
communicate using intervals rather than point values
the interval widens as uncertainty increases.

Check to track your progress (stored locally)

Learning progress0%

account_treeGenealogyexpand_more

menu_bookReferencesexpand_more

Source support: Explicit

link
zh.wikipedia.orghttps://zh.wikipedia.org/wiki/%E5%8D%80%E9%96%93%E4%BC%B0%E8%A8%88ZH · Explicit
verified

RELATED MODELS