Higher-order spline outperforms Wall Street model in forecasting China’s implied volatility
image: We can see the difference between SVI and spline more clearly here. As expected, SVI curves show nice “smiles.” On the other hand, the spline follows the datapoints more closely but can go only as far as the observed data. In all four graphs for the four kinds of options, spline does fit the market implied volatilities better. In accordance with previous discussions, SVI fits the observed implied volatilities rather poorly for CSI300 index options.
Credit: Yuhan Jiao (Tianjin University of Finance and Economics, China) Shuxin Guo (Southwest Jiaotong University, China) Qiang Liu(Southwestern University of Finance and Economics, China)
Background and Motivation
As China's options market experiences rapid growth yet remains relatively young and under-studied compared to established Western markets, understanding which volatility models perform best locally has become a pressing question. China Finance Review International (CFRI) brings you a study titled “Implied volatility modelling and forecasting: evidence from China”, which investigates the performance of different approaches to modelling and forecasting implied volatility in China’s emerging options market. As China’s options trading grows rapidly, the paper addresses a critical gap: which volatility models—industry-popular parametric ones or academically favoured nonparametric ones—perform better in this unique market context.
Methodology and Scope
The research compares two representative models: the parametric Stochastic Volatility Inspired (SVI) model, widely used on Wall Street, and the nonparametric *5th-order spline interpolation*, often used in academia to extract risk-neutral densities. The study also includes two interpolating spline methods for volatility forecasting. Using daily data from 2021, the authors analyse four key Chinese options: China 50 ETF options, CSI 300 index options, Soybean meal futures options, Copper futures options. Performance is evaluated using in-sample and out-of-sample pricing errors, volatility arbitrage profit/loss, and delta-hedging effectiveness.
Key Findings and Contributions
- The 5th-order spline consistently outperforms SVI in both modelling and forecasting implied volatility across all four options.
- SVI, though popular abroad, shows higher pricing errors and is less robust in calibration, especially for longer maturities.
- Spline models also lead to smaller arbitrage profits, suggesting they align more closely with actual market pricing.
- The study is the first to apply the 5th-order spline for volatility modelling (not just density extraction) and to systematically compare it with SVI in China.
Why It Matters
The findings highlight a disconnect between global practices and local market behaviour. Despite SVI’s popularity in the U.S., it is not well-suited to China’s less experienced options market. The results suggest that simpler, more flexible nonparametric methods may be more effective in emerging markets where trading patterns and data structures differ.
Practical Applications
- Traders and market makers in China can use the 5th-order spline for more accurate pricing and hedging.
- The study provides a volatility arbitrage strategy based on model deviations, offering a practical tool for identifying trading opportunities.
- Financial institutions and derivatives exchanges can adopt spline-based models for volatility surface construction and risk management.
Discover high-quality academic insights in finance from this article published in China Finance Review International. Click the DOI below to read the full-text original! Open access for a limited time!