{"paper_id":"2404.18761","title":"A pure dual approach for hedging Bermudan options","authors":["Aurélien Alfonsi","Ahmed Kebaier","Jérôme Lelong"],"paper_date":"2024-04-29","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2404.18761v2","page_url":"https://thequant.space/flowcharts/a-pure-dual-approach-for-hedging-bermudan-options/","scores":{"empirical_rigor":7,"math_complexity":8.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a new purely dual algorithm to compute the hedging portfolio and initial value of Bermudan options using a Monte Carlo least‑square approach.","quote":"This paper develops a new dual approach to compute the hedging portfolio of a Bermudan option and its initial value. It gives a “purely dual” algorithm following the spirit of Rogers [2010] in the sense that it only relies on the dual pricing formula.","summary":"This paper develops a new dual approach to compute the hedging portfolio of a Bermudan option and its initial value. It gives a “purely dual” algorithm following the spirit of Rogers (2010) in the sense that it only relies on the dual pricing formula. The key is to rewrite the dual formula as an…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The sub‑vector spaces H_n are spanned by L random variables ΔX_{n,ℓ} and L does not depend on n.","type":"model","stated":"explicit","quote":"Here, L does not depend on n for simplicity, and we note α·ΔX_n ="},{"assumption":"The increments ΔX_i and the variables Z_i have finite moments of order 4.","type":"statistical","stated":"explicit","quote":"If we assume moreover that ΔX_i and Z_i have finite moments of order 4, then √Q(α^Q_n−α_n)"},{"assumption":"Monte Carlo paths are independent and the number of paths Q is large enough for the law of large numbers to apply.","type":"execution","stated":"explicit","quote":"we have Q independent paths of the underlying process Z^q_n and of the martingale increments ΔX^q_n"},{"assumption":"Rebalancing can be performed at the prescribed discrete dates without additional market impact.","type":"execution","stated":"implied","quote":"The higher is the frequency, the closer is the estimated dual price to the primal price given by the Longstaff and Schwartz algorithm."}]},"datasets":{"status":"available","items":[{"name":"Synthetic Bermudan put option on a single asset","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"Table 1: Prices for a put option using a basis of P local functions with K = S0 = 100, T = 0.5, r = 0.06, σ = 0.4 and N = 10 exercising dates."},{"name":"Synthetic Bermudan min‑put on two assets","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"Results are given in Table 5 for the local basis regression and in Table 6 for the polynomial basis."},{"name":"Synthetic Bermudan basket put in dimension 3","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"Table 7: Prices for a basket put option in dimension d = 3 using a basis of local functions with K = S0 = 100, T = 1, r = 0.05, σ_i = 0.2, ρ = 0.3 and 10 exercising dates."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"synthetic"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"no","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"pricing-calibration","findings":[{"finding":"The algorithm converges and Monte Carlo estimators of the hedging coefficients are almost surely consistent.","kind":"theoretical","quote":"α^Q_n →_{Q→∞} α_n, almost surely, where α_n is defined by (20)."},{"finding":"Prices obtained with the dual algorithm are very close to Longstaff‑Schwartz prices, e.g., 22.82 vs 22.83 for the min‑put example.","kind":"pricing-calibration","quote":"We obtain in Table 5 a price of 22.82, which is very close to 22.83 obtained in [Rogers, 2002, Table 4.5]."},{"finding":"Including European options in the hedging portfolio significantly improves the price and reduces P&L variance (e.g., variance reduced from 36.6 to 4).","kind":"performance","quote":"adding European options in the hedging strategy does not only improve the selling price but also reduces the variance of the P&L. The variance is reduced from 36.6 to 4."},{"finding":"Higher rebalancing frequency narrows the gap between dual and primal prices, but the benefit diminishes when European options are included.","kind":"performance","quote":"The higher is the frequency, the closer is the estimated dual price to the primal price given by the Longstaff and Schwartz algorithm"},{"finding":"The dual algorithm yields a more peaked P&L distribution than a static delta‑hedging approach, indicating a better hedge.","kind":"performance","quote":"the left plot is more peaked than the right one, which indicates a better hedge provided by our algorithm."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":26,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":1},"independent_papers":[],"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":1,"extracted_at":"2026-11-14T18:10:23Z"},"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/"}