Interval-Based Abnormality Detection and Event Returns in Daily Stock Data

Fuente: Zenodo
Enregistré dans:
Détails bibliographiques
Auteur principal: Burk, Kevin
Format: Recurso digital
Langue:anglais
Publié: Zenodo 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866901210470023168
author Burk, Kevin
author_facet Burk, Kevin
contents <p>This record contains the preregistration for the study <em>Interval-Based Abnormality Detection and Event Returns in Daily Stock Data</em>. The document specifies all methodological decisions made prior to analysis, including the forecast-generation process, interval construction, abnormality definitions, BUY/SELL event rules, portfolio mechanics, outcome measures, statistical tests, and all figures to be produced.</p> <p>The preregistration fixes a 10-year evaluation window (2015–2024) using daily forecasts constructed from historical data beginning in 2005. The study evaluates the economic characteristics of interval-defined abnormality events across 30 U.S. equities, using a deterministic, non-adaptive, multi-model forecasting approach.</p> <p>The preregistration freezes all analytic choices in advance of data analysis. The full protocol is provided in the attached PDF.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_17644819
institution Zenodo
language eng
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle Interval-Based Abnormality Detection and Event Returns in Daily Stock Data
Burk, Kevin
abnormality detection
interval forecasting
empirical residuals
daily stock data
event returns
nonparametric analysis
backtesting
forecasting evaluation
financial time series
preregistered study
<p>This record contains the preregistration for the study <em>Interval-Based Abnormality Detection and Event Returns in Daily Stock Data</em>. The document specifies all methodological decisions made prior to analysis, including the forecast-generation process, interval construction, abnormality definitions, BUY/SELL event rules, portfolio mechanics, outcome measures, statistical tests, and all figures to be produced.</p> <p>The preregistration fixes a 10-year evaluation window (2015–2024) using daily forecasts constructed from historical data beginning in 2005. The study evaluates the economic characteristics of interval-defined abnormality events across 30 U.S. equities, using a deterministic, non-adaptive, multi-model forecasting approach.</p> <p>The preregistration freezes all analytic choices in advance of data analysis. The full protocol is provided in the attached PDF.</p>
title Interval-Based Abnormality Detection and Event Returns in Daily Stock Data
topic abnormality detection
interval forecasting
empirical residuals
daily stock data
event returns
nonparametric analysis
backtesting
forecasting evaluation
financial time series
preregistered study
url https://doi.org/10.5281/zenodo.17644819