{"paper_id":"2609.19102","title":"Optimal entry and exit for variance swaps: closed-form rules for the perpetual contract","authors":["Jun Maeda"],"paper_date":"2026-09-16","paper_type":"theoretical","paper_url":"https://arxiv.org/pdf/2609.19102v1","page_url":"https://thequant.space/flowcharts/optimal-entry-and-exit-for-variance-swaps-closed-form-rules-for-the-perpetual/","scores":{"empirical_rigor":4,"math_complexity":9,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper derives closed‑form entry and exit thresholds for a perpetual, continuously settled variance swap, showing each threshold is the unique root of a smooth‑pasting equation in confluent hypergeometric functions.","quote":"We ask when a trader should put such a position on and when she should take it off, and for a perpetual, continuously settled contract we answer both in closed form: each threshold is the unique root of a smooth-pasting equation in confluent hypergeometric functions. This paper solves that problem in closed form for a perpetual, continuously settled variance swap: both the level at which the position is opened and","summary":"Variance swaps are a convenient instrument for trading vega and convexity, and a listed contract now trades on Cboe. We ask when a trader should put such a position on and when she should take it off, and for a perpetual, continuously settled contract we answer both in closed form: each threshold…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The underlying variance follows the Heston stochastic volatility model.","type":"model","stated":"explicit","quote":"1.2 The Heston model We work throughout in the Heston model.","support":"full"},{"assumption":"The contract is perpetual and continuously settled, so there is no fixed maturity.","type":"other","stated":"explicit","quote":"We ask when a trader should put such a position on and when she should take it off, and for a perpetual, continuously settled contract we answer both in closed form: each threshold is the unique root of a smooth-pasting equation in confluent hypergeometric functions. This paper solves that problem in closed form for a perpetual, continuously settled variance swap: both the level at which the position is opened and","support":"full"},{"assumption":"Under the physical measure P there is a variance risk premium that makes the stopping problem non‑trivial.","type":"market","stated":"explicit","quote":"Under the physical measure with a variance risk premium it becomes meaningful, and then reduces: the accrued variance separates exactly, the maturity, strike and costs are absorbed into a single forcing term whose sign fixes the geometry of the exercise region, and what is left on the perpetual is an affine reward on a CIR process, which the optimal-stopping literature already solves. The problem must be posed under","support":"full"},{"assumption":"Mark‑to‑market values are a Q‑martingale, so the pricing measure problem has no content.","type":"model","stated":"explicit","quote":"Under the pricing measure the question has no content, the mark-to-market being a martingale. The problem must be posed under P. Under the pricing measure it has no content: the mark-to-market is a Q-martingale (Lemma 2.1), the discount factor applied to the stopping payoff cancels against the one inside it (Proposition 2.2), and every stopping time attains the supremum (Corollary 2.3).","support":"full"},{"assumption":"Transaction costs are represented by a constant carrying charge cm (e.g., 0.004 per year) and a spread on entry/exit.","type":"execution","stated":"explicit","quote":"For the perpetual we take the tear-up concessions ¯s = ¯se = 0.002, so that closing a perpetual costs what closing a one-year dated swap costs; Remark 6.2 reports what happens if the two are instead tied together at ¯s = s/δ. Where a carrying charge is needed we use cm = 0.004 per year.","support":"full"},{"assumption":"No discounting is applied to the perpetual reward (funding enters as a running charge).","type":"model","stated":"explicit","quote":"This needs the absence of discounting that Proposition 2.2 supplies, so funding enters as a running charge rather than a discount rate.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Cboe listed S&P 500 Variance Futures contract","provider":"Cboe","kind":"market-prices","access":"public","period":null,"universe":"S&P 500 index variance futures","quote":"The question has become more than academic: Cboe listed a redesigned S&P 500 Variance Futures contract on 23 September 2024 (Cboe, 2024a; Bartholomew, 2024), so a position can now be entered and unwound on exchange, with margin, at a spread narrower than the historical OTC market offered.","support":"full","known_id":"cboe"},{"name":"Option prices used for VIX calculation","provider":"Cboe","kind":"options","access":"public","period":null,"universe":"S&P 500 options (first two monthly expiries)","quote":"Equation (5) is the foundation of the VIX calculation, which selects sufficiently liquid options in the first two monthly expiries and evaluates a discrete version of (5); see Cboe (2024b).","support":"full","known_id":"cboe"}],"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":"R4a","candidates":{"front_matter":"n/a","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"A constant carrying charge cm = 0.004 per year is used, and the exit rule incorporates the observable trading spread.","card_quote":"For the perpetual we take the tear-up concessions ¯s = ¯se = 0.002, so that closing a perpetual costs what closing a one-year dated swap costs; Remark 6.2 reports what happens if the two are instead tied together at ¯s = s/δ. Where a carrying charge is needed we use cm = 0.004 per year.","meaning":"the paper reports no trading or execution result; the cost the extractor found is a model parameter"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"Both entry and exit thresholds are unique roots of smooth‑pasting equations in confluent hypergeometric functions, giving a closed‑form solution.","kind":"theoretical","quote":"We ask when a trader should put such a position on and when she should take it off, and for a perpetual, continuously settled contract we answer both in closed form: each threshold is the unique root of a smooth-pasting equation in confluent hypergeometric functions. This paper solves that problem in closed form for a perpetual, continuously settled variance swap: both the level at which the position is opened and","support":"full"},{"finding":"The exit rule is directly observable from the variance risk premium and the trading spread.","kind":"theoretical","quote":"An exit rule follows from the premium and the trading spread, both observable. Opening the position and closing it are not mirror images.","support":"full"},{"finding":"For the short side, an entry rule exists only for a narrow interval of carrying charges and triggers deep in the upper tail of the physical variance distribution (around the 95th percentile).","kind":"theoretical","quote":"For the short --- the only side worth opening under the empirical sign of the premium --- an entry rule exists only for an interval of carrying charges, and even there triggers only deep in the upper tail of the physical law: a trader who is out of the market pays nothing to stay out, so an operational entry rule needs a cost of idle capital that the exit rule does not. For the short — the only side worth opening","support":"partial"},{"finding":"Charging idle‑capital cost (e.g., 50 basis points per year) shifts the entry threshold from the 95th to the 58th percentile of the stationary law.","kind":"theoretical","quote":"Charging idle capital fifty basis points a year moves it to the 58th.","support":"partial"},{"finding":"The reduction from the original variance‑swap problem to an affine reward on a CIR process is the main methodological contribution; the optimal‑stopping machinery itself is borrowed from existing literature.","kind":"theoretical","quote":"The reduced problem, an affine reward on a CIR process entered and exited once against transaction costs, is one Dayanik and Karatzas (2003) and Leung et al. (2014) have already solved; the work is in showing that a variance swap gives rise to a problem of that form, and in reading what the resulting thresholds say. 1. Sections 5– 7 use their machinery, and we claim no novelty for it: our contribution is the","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":18,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"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":2,"extracted_at":"2026-11-14T19:39:49Z"},"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/"}