Follow us on LinkedIn
Value at Risk (VaR) is a widely used risk metric in finance, providing an estimate of the potential loss on a portfolio or investment over a specific time horizon, under normal market conditions. It quantifies the level of financial risk within a certain confidence interval, expressing the maximum loss that might be incurred over a given period.
Calculating VaR for linear instruments is straightforward, but for instruments with convexities, such as options, the process becomes more complex. Reference [1] has proposed a method for calculating the daily VaR of an options portfolio. The authors stated,
In our proposed method, we initiate the process by extracting the volatility surface for each day from option prices. Subsequently, volatility shocks are computed based on historical data on the volatility surface, resulting in a database of projected volatility surfaces. Simultaneously, employing Monte Carlo simulation, we generate next-day prices for the underlying asset. These volatility shocks and next-day prices are then input into the Black-Scholes formula pricing model to obtain forecasts for option prices the following day.
Numerical tests reveal the superior performance of the PSP method, particularly at high confidence levels (e.g., 95%). The DM test emphasizes the effectiveness of localized volatility models over reference volatility, capturing market conditions more accurately. Furthermore, incorporating volatility, regardless of its form (reference or localized), demonstrates potential improvements in forecasting accuracy.
The method essentially begins by calculating daily stock moves using historical data. Subsequently, it determines the distribution of implied volatility associated with each simulated stock price through a method known as Parametric Surface Projection. This generated set of implied volatilities is then utilized to calculate the daily VaR of an options portfolio.
We find this method interesting and think it’s a good complement to analytical methods such as the delta-gamma-vega approach.
Let us know what you think in the comments below or in the discussion forum.
References
[1] Shiva Zamani, Alireza Moslemi Haghighi, Hamid Arian, Temporal Volatility Surface Projection: Parametric Surface Projection Method for Derivatives Portfolio Risk Management, https://arxiv.org/abs/2311.14985v1
Further questions
What's your question? Ask it in the discussion forum
Have an answer to the questions below? Post it here or in the forum
JPMorgan Chase will be watched closely for clues on how banks fared at the start of the year.
From new shows like "Fallout" and "The Sympathizer" to true crime docs like "What Jennifer Did," here are all the best things to watch this weekend.
BlackRock saw its assets under management grow by $1.4 trillion in 12 months as client funds poured in and markets performed well.
Jaime Muñoz left the Silicon Valley start-up hub for a fintech job in Chicago, hoping to get in early at a company and build wealth.
TORONTO — VerticalScope Holdings Inc. (“VerticalScope” or the “Company”) (TSX: FORA) announced today that the Company’s first quarter 2024 financial results will be released after the market close on Tuesday, May 7, 2024. Management will host a conference call and webcast to discuss the Company’s…