Testing Whether Volatility Model Gains Persist: A Prespecified Holdout in Tail Risk Forecasting

Testing eight volatility models on equity indices with prespecified holdout periods, the research finds that gains from more complex models often do not persist across markets or time.

Read the paper PDF

View as Markdown Open in Claude Open in ChatGPT Open in PerplexitySite index for AI (llms.txt)

Featured in No. 133 on 2 Oct 2026 · 2 days after release

Mean FZ0 differences across volatility models with confidence intervals, multiple markets and specifications. — Figure 1: Main-OOS matched specification contrasts. Points report the mean FZ0 difference for the model listed first minus its benchmark; vertical bars are ±1.96 paired HAC standard errors. Negative differences favour the model listed first. The panels show all six prespecified contrasts for five m…

Released: 30 Sep 2026
First featured: No. 133 · 2 Oct 2026
Published in: Not yet, as far as Semantic Scholar knows
Fanfare: 3 of 5
Identifier: RePEc:cte:wsrepe:50798
Authors: Honfei Guo et al.

Citations and venue from Semantic Scholar (ODC-BY), refreshed weekly. Summary: Quant Letter (CC BY 4.0).

Related papers

A Risk Sensitive Contract-unified Reinforcement Learning Approach for Option Hedging

Combining Deep Learning and GARCH Models for Financial Volatility and Risk Forecasting

Tail Risk Hedging in Equity Portfolios: Evaluating the Cost-Effectiveness of Out-of-the-Money Put Options versus Dynamic Tail-Risk Mitigation Strategies During High-Volatility Regimes

Tail-Risk Forecasting with General Cubic Distributions

Deep Hedging to Manage Tail Risk

Multivariate Affine GARCH with Heavy Tails: A Unified Framework for Portfolio Optimization and Option Valuation