{"paper_id":"2508.09863","title":"Marketron Through the Looking Glass: From Equity Dynamics to Option Pricing in Incomplete Markets","authors":["Igor Halperin","Andrey Itkin"],"paper_date":"2025-08-13","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2508.09863v2","page_url":"https://thequant.space/flowcharts/marketron-through-the-looking-glass-from-equity-dynamics-to/","scores":{"empirical_rigor":6,"math_complexity":9,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a utility‑based option pricing framework for the Marketron model, solved via a novel operator‑splitting and Gaussian RBF method that reduces the high‑dimensional PDE to tractable integral equations.","quote":"We develop a utility-based pricing approach that constructs a risk-adjusted measure via the dual solution of an optimal investment problem.","summary":"The Marketron model, introduced by [“Halperin, Itkin, 2025”], describes price formation in inelastic markets as the nonlinear diffusion of a quasiparticle (the marketron) in a multidimensional space comprising the log-price x, a memory variable y encoding past money flows, and unobservable return…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Constant interest rate is assumed.","type":"model","stated":"explicit","quote":"For simplicity, we assume a constant interest rate, i.e., rt = r."},{"assumption":"Constant volatilities are assumed for the stochastic terms.","type":"model","stated":"explicit","quote":"given our model’s assumption of constant volatilities, as the Green’s function for the homogeneous problem has a known analytical form"},{"assumption":"The utility function U is concave, strictly increasing and differentiable.","type":"model","stated":"explicit","quote":"Let U : R →R ∪{−∞} be a utility function, assumed to be a concave, strictly increasing and differentiable function."},{"assumption":"Memory variable y_t and signals θ_t are non‑tradable hidden stochastic variables, creating market incompleteness.","type":"market","stated":"explicit","quote":"the memory variable yt and signals θt are non‑tradable, hidden stochastic variables."},{"assumption":"All stochastic terms are Gaussian, even though the overall model is non‑Gaussian due to nonlinear drifts.","type":"statistical","stated":"explicit","quote":"Since all stochastic terms are Gaussian (though the model itself is non‑Gaussian due to nonlinear drifts)"}]},"datasets":{"status":"available","items":[{"name":"S&P 500 return time‑series","provider":"S&P 500 index","kind":"market-prices","access":"public","period":null,"universe":"S&P 500 constituents","quote":"In [Halperin and Itkin, 2025], we calibrated the 3D version of the model to S&P 500 return time-series using particle filtering and optimization methods."},{"name":"SPX options prices","provider":"S&P 500 index options market","kind":"options","access":"public","period":null,"universe":"SPX options","quote":"While this calibration approach proved relatively successful, alternative market data could be used, such as S&P500 index options"},{"name":"Simulated log‑returns from calibrated Marketron model","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"Additionally, in the subsequent analysis, we aim to examine the underlying stock log‑returns generated by simulating our model using parameters calibrated to opti…"}],"known":[{"id":"optionmetrics","name":"OptionMetrics IvyDB","provider":"OptionMetrics","kind":"options","access":"commercial","quote":"For calibration, we use daily OptionMetrics snapshots of Call and Put prices for a given maturity T. We restrict strikes K to S/K < 1.05 for Puts and K/S < 1.05 for Calls, including OTM, ATM, and near-ATM ITM options.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"n/a","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"pricing-calibration","findings":[{"finding":"Calibration to equity data yields an annualized default intensity of about 18 basis points, matching typical market‑implied values of 10–50 bps.","kind":"pricing-calibration","quote":"The calibrated model yields an annualized default intensity of approximately 18 basis points, aligning well with market‑implied default intensities typically ranging from 10 to 50 basis points."},{"finding":"When calibrated to SPX options, the model reproduces option smiles but consistently overestimates underlying volatility compared to the time‑series.","kind":"statistical","quote":"The main discrepancy is about volatility - the model consistently overestimates volatility compared to"},{"finding":"The proposed RBF‑based solver requires only about 5 collocation points in the y and θ dimensions, far fewer than finite‑difference methods.","kind":"theoretical","quote":"The key advantage of the RBF method is that it requires significantly fewer collocation points in the y and θ dimensions — approximately 5 points suffice, whereas FD methods typically need an order of magnitude more points for comparable accuracy."},{"finding":"Operator‑splitting combined with Cole‑Hopf transformation converts the nonlinear PDE into linear equations that can be solved efficiently via Gaussian RBFs.","kind":"theoretical","quote":"We demonstrate that the nonlinear equations can be transformed into linear counterparts using the Cole‑Hopf transformation."},{"finding":"Pricing multiple strikes can be done in a single linear solve because only the source term depends on the strike.","kind":"performance","quote":"the only part of Eq. (44) which depends on the strike K is I(τ, x). Therefore, linear systems of equations that appear in our RBF method having various right‑hand sides, can be solved simultaneously."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":44,"citations":{"cited_by_in_corpus":3,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":3},"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":"2026-11-14T17:29:53Z"},"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/"}