Revisiting the Idiosyncratic Volatility Puzzle in China: Evidence from Deep Learning Models
“We show that deep learning approaches, such as the Temporal Fusion Transformer (TFT), and conventional time-series models like ARMA(1,1) substantially outperform the martingale benchmark in predictive accuracy.”
AssetEquities
StrategyLow-volatility anomaly
TechniqueTransformerfeedforward network
GeographyEuropeChina
Also stated: Intraday, Excess return
Missing from abstract: benchmark, out-of-sample test, costs, multiple testing, universe
Institutional-grade audit of this strategy.
£5 — Request the audit- Audit of the paper claims accounting for costs, luck, crises, originality, geography, etc.
- Reproduction code integratable into Bloomberg, IBKR, etc.
- Bias-free datasets for backtesting included
- Delivered in 2 working days
The paper's neighbourhoodLow-volatility anomaly by year
FamilyLow-volatility anomaly25Momentum and trend following289Factor investing240Machine learning return prediction15318 more · 977
AssetEquities21
The family is counted across the whole universe. Asset, method and authors are counted inside the family.
| Similar papers | citation | same method | same asset | same authors | |
|---|---|---|---|---|---|
| Correlation-Gated Risk Budgeting for Equity Factor Portfolios by Guo | 2026 | ||||
| From the White House to Wall Street: How U.S. Political Regimes and Policy Uncertainty Shape the Low-Risk Anomaly by Velinova | 2025 | ||||
| The Volatility Puzzle of the Beta Anomaly by Barroso et al. | 2021 | ||||
| The Volatility Effect in China by Blitz et al. | 2021 | ||||
| The Low-volatility Anomaly and the Adaptive Multi-Factor Model by Jarrow et al. | 2021 |