{"paper_id":"2309.00540","title":"Instabilities of Super-Time-Stepping Methods on the Heston Stochastic Volatility Model","authors":["Fabien Le Floc'h"],"paper_date":"2023-09-01","paper_type":"theoretical","paper_url":"https://arxiv.org/pdf/2309.00540v1","page_url":"https://thequant.space/flowcharts/instabilities-of-super-time-stepping-methods-on-the-heston-s/","scores":{"empirical_rigor":6,"math_complexity":8.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper investigates instabilities of explicit super‑time‑stepping schemes, such as Runge‑Kutta‑Chebyshev (RKC) and Runge‑Kutta‑Legendre (RKL), when applied to the Heston stochastic volatility PDE and to a barrier option under the Black‑Scholes model.","quote":"This note explores in more details instabilities of explicit super-time-stepping schemes, such as the Runge-Kutta-Chebyshev or Runge-Kutta-Legendre schemes, noticed in the litterature, when applied to the Heston stochastic volatility model. The super-time-stepping schemes allow to circumvent this limitation.","summary":"This note explores in more details instabilities of explicit super-time-stepping schemes, such as the Runge-Kutta-Chebyshev or Runge-Kutta-Legendre schemes, noticed in the litterature, when applied to the Heston stochastic volatility model. The stability remarks are relevant beyond the scope of…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The underlying asset follows the Heston stochastic volatility model.","type":"model","stated":"explicit","quote":"This note explores in more details instabilities of explicit super-time-stepping schemes, such as the Runge-Kutta-Chebyshev or Runge-Kutta-Legendre schemes, noticed in the litterature, when applied to the Heston stochastic volatility model.","support":"full"},{"assumption":"A low vol‑of‑variance parameter (σ = 4%) is used for the Heston experiments.","type":"model","stated":"explicit","quote":"Foulon and In’t Hout [2010] noticed instabilities of the Runge-Kutta-Chebyshev (RKC) super-time-stepping scheme with relatively large shift (ϵ = 10) on the Heston PDE in the context of a low vol-of-variance parameter (σ = 4%).","support":"full"},{"assumption":"The barrier option payoff is a manufactured contract that pays $1 if the spot price at maturity lies between 10 and 100.","type":"other","stated":"explicit","quote":"We further consider an expiry barrier option which pays $1 if the asset spot price X (T ) at maturity T = 1 year is between 10 and 100 and zero otherwise, assuming X (0) = 100.","support":"full"},{"assumption":"Central differencing is used for spatial discretisation, with non‑uniform grids as in prior work.","type":"model","stated":"explicit","quote":"We discretize the PDE with central differencing, except at the boundaries xmin = 0 and xmax = 150 on a non-uniform grid as in [Le Floc’h 2014].","support":"full"},{"assumption":"The explicit Euler scheme has a limited stability region and would require many time‑steps for advection‑diffusion PDEs.","type":"model","stated":"explicit","quote":"On the other side, the explicit Euler scheme, with its limited stability region, is well known to require too many time-steps to be practical for advection-diffusion PDEs.","support":"full"},{"assumption":"The price is assumed linear in the variance dimension at the upper variance boundary and linear in the asset dimension at the spatial boundaries.","type":"model","stated":"explicit","quote":"At v = vmax, Le Floc’h and Oosterlee [2019] follow Andersen and Piterbarg [2010, p. 385-386] and let the price be linear in the variance dimension: ∂2 f ∂v2 (x,v,t) = ∂2 f ∂x∂v (x,v,t) = 0, (3) for x ∈(xmin,xmax). At xmax and xmin, we consider that the value is linear along x, which leads to ∂2 f ∂x2 (x,v,t) = ∂f ∂v (x,v,t) = 0, (4) for v ∈[vmin,vmax] and x ∈{xmin,xmax}.","support":"full"}]},"datasets":{"status":"none_found","items":[],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"Instabilities were observed for the RKC super‑time‑stepping scheme with a large damping shift (ε = 10) on the Heston PDE when the vol‑of‑variance is low (σ = 4%).","kind":"negative","quote":"Foulon and In’t Hout [2010] noticed instabilities of the Runge-Kutta-Chebyshev (RKC) super-time-stepping scheme with relatively large shift (ϵ = 10) on the Heston PDE in the context of a low vol-of-variance parameter (σ = 4%). This is somewhat surprising, since the large shift increases the stability and damping properties of the scheme significantly [Verwer et al. 2004].","support":"full"},{"finding":"The RKL scheme exhibits spurious oscillations around the strike in the barrier option price under Black‑Scholes parameters (σ = 2%, r = 10%).","kind":"negative","quote":"The option price computed with RKL scheme presents additional oscillations, (a) RKL. (b) RKG or TR-BDF2. (c) Eigenvalues. Figure 4. Spurious oscillation around K appear with the RKL scheme.","support":"partial"},{"finding":"Oscillations with RKL diminish when the number of time‑steps is increased (e.g., from l = 20 to l = 50) or when exponential fitting is applied.","kind":"performance","quote":"Small oscillations are visible with the RKL scheme and a low number of time-steps such as l = 20. They disappear if we increase the number of time-steps (for example to l = 50) and are absent with the RKG and TR-BDF2 schemes.","support":"full"},{"finding":"RKG and TR‑BDF2 schemes do not show the oscillatory behavior observed with RKL in the same settings.","kind":"performance","quote":"The option price computed with RKL scheme presents additional oscillations, (a) RKL. (b) RKG or TR-BDF2. (c) Eigenvalues. Figure 4. Partial exponential fitting is used not present with the TR-BDF2 or RKG schemes, without exponential fitting.","support":"full"},{"finding":"A more careful discretisation that reduces point concentration near v = 0 helps avoid the observed oscillations.","kind":"theoretical","quote":"A more careful discretization (less extreme concentration of points near v = 0) would also help to avoid those oscillations.","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[],"contradicts":[],"qualifies":[],"compared_with":23,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":1,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[{"paper":{"paper_id":"2512.19611","title":"Heston vol-of-vol and the VVIX","paper_date":"2025-12-22","authors":["Jherek Healy"],"empirical_rigor":3,"page_url":"https://thequant.space/flowcharts/heston-vol-of-vol-and-the-vvix/"},"how":"uses_method_or_data","quote":null}],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":2,"extracted_at":"2026-11-14T20:01:27Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}