Testing Whether Volatility is Local or Stochastic
The paper develops and tests a high-frequency specification test for local versus stochastic volatility in asset prices.
What it examines
This study develops a test to decide whether a financial asset's volatility can be modeled simply as a function of its level (local volatility) or if a more complex stochastic model is needed. It uses high frequency data methods, nonparametric estimation, and techniques to handle noise and possible jumps in the data.
What it concludes
The results show that simple local volatility models are rejected for various financial series, implying the need for more complex models. These methods can help in pricing, risk management, and model specification in finance, with future work focusing on other asset classes and refining techniques.
Evidence objects
A new study investigates whether local volatility models suffice or dynamic stochastic volatility models are essential, using high-frequency data and robust econometric tests comparing observational data estimators across financial instruments.
key_findings bullet 1 · key_findings · validation V0
Surprising evidence reveals that simpler local models, despite common usage, fail to effectively capture intricate volatility patterns, especially in markets characterized by jumps and microstructure noise, necessitating dynamic stochastic approaches.
key_findings bullet 2 · key_findings · validation V0
The authors introduce novel definitions and statistical methods, including noise-robust estimators and pre-averaging techniques, and use derivations, Monte Carlo simulations, and market data while acknowledging limited testing in market conditions.
key_findings bullet 3 · key_findings · validation V0
This paper introduces a novel testing framework to differentiate between local and stochastic $\sigma$ models. Employing high-frequency asymptotics while addressing jumps and microstructure noise, it provides robust empirical insights that advance derivative pricing and financial econometrics. Its original methodologies present perspectives that captivate quantitative experts and enrich volatility modeling literature.
key_findings bullet 4 · key_findings · validation V0
Raw abstract and provenance
- There is broad empirical agreement that the volatility of most economic and financial variables is time-varying. One possible way to capture this in continuoustime models …
Source row: 1911 · abstract type: snippet