Learning Rare Events: Deep Learning Approaches to Extreme Price Prediction — A Systematic Review Protocol

Read full text Summarise with VIRAT
Authors: Mark Sinclair, Andrew Shepley, Farshid Hajati
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This registration describes a systematic literature review protocol investigating deep learning approaches to forward-looking price spike prediction across multiple market domains, including electricity, financial, commodity, and cryptocurrency markets. The review focuses specifically on rare-event forecasting, where extreme price movements represent low-frequency, high-impact outcomes characterised by non-stationarity, heavy-tailed distributions, and class imbalance. The objective is to synthesise modelling approaches, spike definitions, task formulations, and evaluation practices, and to identify methodological gaps in current research. The review follows PRISMA guidelines and employs a structured search, screening, and narrative synthesis process.

Fields of Study & Subjects

Stock Market Forecasting Methods Energy Load and Power Forecasting Forecasting Techniques and Applications

Publication details

Publication date
1 January 2026
Language
EN
Document type
Preprint