Saved in:
Bibliographic Details
Main Author: Lehmann, Niklas V.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.07575
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915661144391680
author Lehmann, Niklas V.
author_facet Lehmann, Niklas V.
contents When a dataset contains forecasts on unscheduled events, such as natural catastrophes, outcomes may be censored or ``hidden'' since some events have not yet occurred. This article finds that this can lead to a selection bias which affects the perceived accuracy and calibration of forecasts. This selection bias can be eliminated by excluding forecasts on outcomes which have been verified surprisingly early.
format Preprint
id arxiv_https___arxiv_org_abs_2512_07575
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Surprisingly-early bias in forecasts for unscheduled events
Lehmann, Niklas V.
Methodology
When a dataset contains forecasts on unscheduled events, such as natural catastrophes, outcomes may be censored or ``hidden'' since some events have not yet occurred. This article finds that this can lead to a selection bias which affects the perceived accuracy and calibration of forecasts. This selection bias can be eliminated by excluding forecasts on outcomes which have been verified surprisingly early.
title Surprisingly-early bias in forecasts for unscheduled events
topic Methodology
url https://arxiv.org/abs/2512.07575