Paper: arXiv 2309.07023
Authors: Christian Bayer, Simon Breneis
Abstract
The rough Heston model is a very popular recent model in mathematical finance; however, the lack of Markov and semimartingale properties poses significant challenges in both theory and practice. A way to resolve this problem is to use Markovian approximations of the model. Several previous works have shown that these approximations can be very accurate even when the number of additional factors is very low. Existing error analysis is largely based on the strong error, corresponding to the \(L^2\) distance between the kernels. Extending earlier results by [“Abi Jaber and El Euch, SIAM Journal on Financial Mathematics 10(2):309–349, 2019”], we show that the weak error of the Markovian approximations can be bounded using the \(L^1\)-error in the kernel approximation for general classes of payoff functions for European style options. Moreover, we give specific Markovian approximations which converge super-polynomially in the number of dimensions, and illustrate their numerical superiority in option pricing compared to previously existing approximations. The new approximations also work for the hyper-rough case \(H > -1/2\).
Complexity vs Empirical Score
- Math Complexity: 9.0/10
- Empirical Rigor: 3.0/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper is heavily mathematical, focusing on theoretical error bounds using advanced stochastic calculus, fractional calculus, and PDE/Riccati equations for weak error analysis, with almost no mention of backtesting or implementation details.
Research Flowchart
flowchart TD A["Research Goal: Analyze Weak Error of Markovian Approximations for Rough Heston"] --> B["Methodology: L^1 Kernel Error Analysis"] A --> C["Input: General Class of European Option Payoffs"] B --> D["Computation: Bound Weak Error via L^1 Distance"] C --> D D --> E["Computation: Construct New Super-Polynomial Approximations"] E --> F["Outcome: Accurate Option Pricing for H > -1/2"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes weak error bounds for Markovian approximations of the rough Heston model using L1‑kernel errors and introduces specific approximations that converge super‑polynomially in the number of dimensions.
Source sentence
we show that the weak error of the Markovian approximations can be bounded using the $L^1$-error in the kernel approximation for general classes of payoff functions for European style options. Moreover, we give specific Markovian approximations which converge super-polynomially in the number of dimensions
Assumes
No explicit assumptions found in the text.
Data
No dataset named in the text.
Costs
Not applicable
Finds 5
- Markovian approximations achieve the same accuracy as the Adams scheme for implied volatility smiles while being dramatically faster (40–270 s vs 6183 s for H = −0.1). performance
Source sentence
the computation of the smile for H = −0.1 with relative accuracy 10−4 took 6183 seconds using the Adams scheme, while the Markovian approximations took 40 – 270 seconds, depending on N and the quadrature rule used
- Quadrature rules based on L1‑error (GG, NGG, OL1, AE) remain accurate for tiny or negative H, whereas L2‑based rules perform poorly. performance
Source sentence
quadrature rules relying on the L2-error (i.e. OL2, AK) perform very badly for tiny H (and are not well-defined for negative H), while quadrature rules relying on the L1-error (i.e. GG, NGG, OL1, AE) still work well even for H = −0.1
- The BL2 quadrature rule, despite optimizing a penalized L2‑error, outperforms all other methods by multiple orders of magnitude even for tiny or negative H. performance
Source sentence
the exception here is BL2, which, despite optimizing the (penalized) L2-error outperforms every other method by multiple magnitudes, even for tiny or negative H
- For sufficiently large N, NGG is expected to outperform GG, though such large N are likely unnecessary for typical applications. theoretical
Source sentence
we guess that for sufficiently large N, NGG will outperform GG, but such large N are likely not necessary for usual applications
- The weak error analysis extends to the hyper‑rough case H > −1/2. theoretical
Source sentence
The new approximations also work for the hyper-rough case $H > -1/2$.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 30 related papers). Coverage is partial, so this is neither replication nor refutation.
9 papers in the archive : Efficient Calibration in the rough Bergomi model by … ; Variance-Optimal Hedging in the Rough Hawkes--Heston Model ; Approximation and regularity results for the Heston model …
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.