📉  Full Distribution

Full Distribution#

Predicting the entire outcome distribution rather than a single value.

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

What it is#

Forecasting the full distribution means predicting the entire predictive distribution — the complete CDF / PDF over all possible outcomes — rather than a summary of it. A point forecast collapses it to one number; a quantile forecast reports a few points; the full distribution keeps everything.

The richest target#

From the full distribution you can derive any summary after the fact — the mean, the median, any quantile, a prediction interval, the probability of exceeding a threshold, or a risk measure such as VaR. Nothing about the uncertainty is discarded.

How it’s judged#

Because it is a whole distribution, it is scored by a rule that reads the entire shape against the outcome — the CRPS, which compares the forecast CDF to the observation’s step CDF — not a point-error metric like MAE.


Theme: Risk & Probabilistic Forecasting  ·  All terminology



See also

Source article Adapted (context, re-expressed) in our own words from: Full Distribution (insightful-data-lab.com).

Tags: purpose: reference topic: terminology level: advanced