Stress testing and scenario analysis are essential tools in portfolio management, helping portfolio and risk managers assess potential vulnerabilities under extreme market conditions. By simulating adverse scenarios such as financial crises, interest rate shocks, or geopolitical events, these techniques provide insights into how a portfolio might behave under stress and identify potential weaknesses.
Reference [1] investigates several stochastic models for simulating stock and volatility paths that can be used in stress testing and scenario analysis. It also proposes a method for evaluating these stochastic models. The models studied include
- Geometric Brownian Motion (GBM),
- Generalized Autoregressive Conditional Heteroskedasticity (GARCH),
- Heston stochastic volatility,
- Stochastic Volatility with Jumps (SVJD), and a novel
- Multi-Scale Volatility with Jumps (MSVJ).
The authors pointed out,
When the objective is to evaluate and simulate scenarios that reflect market crashes, both short-term events and long-term crises, models such as GBM and the Heston model have been shown to be more effective. These models are better equipped to capture the sudden and severe price movements associated with market crashes, as demonstrated by their performance in reproducing historical drawdowns and their ability to capture tail risk…
If the objective is to generate future scenario simulations for option pricing, the MSVJ model has proven to be the most suitable choice. The MSVJ model’s superior performance in capturing the range of the actual TQQQ price, as evidenced by its highest WMCR for both price and volatility, makes it particularly valuable for option pricing…
When the primary goal is to simulate the most realistic price path and volatility paths for TQQQ, the SVJD model has demonstrated superior performance. By capturing both stochastic volatility and jump processes, the SVJD model can generate price and volatility trajectories that closely resemble the observed dynamics of TQQQ. Portfolio managers can utilize this model for more accurate backtesting of trading strategies and better assessment of portfolio risk under various market conditions.
In short, each model has its strengths and weaknesses and serves a particular purpose.
This study is an important contribution to the advancement of portfolio risk management. Let us know what you think in the comments below or in the discussion forum.
References
[1] Kartikay Goyle, Comparative analysis of stochastic models for simulating leveraged ETF price paths, Journal of Mathematics and Modeling in Finance (JMMF) Vol. 5, No. 1, Winter & Spring 2025
Post Source Here: Stock and Volatility Simulation: A Comparative Study of Stochastic Models
source https://harbourfronts.com/stock-volatility-simulation-comparative-study-stochastic-models/
No comments:
Post a Comment