arXiv subjects

Ahmet Umur Özsoy

Publications and source records attributed to Ahmet Umur Özsoy.

4 recordsLinked to original sources

arXiv2025

Selective Forgetting in Option Calibration: An Operator-Theoretic Gauss-Newton Framework

Calibration of option pricing models is routinely repeated as markets evolve, yet modern systems lack an operator for removing data from a calibrated model without full retraining. When quotes become stale, corrupted, or subject to deletion requirements, existing calibration pipelines must rebuild the entire nonlinear least-squares problem, even if only a small subset of data must be excluded. In this work, we introduce a principled framework for selective forgetting (machine unlearning) in parametric option calibration. We provide stability guarantees, perturbation bounds, and show that the proposed operators satisfy local exactness under standard regularity assumptions.

q-fin.MF↗

arXiv2025

The Approach of Sliced Inference in Systems of Stochastic Differential Equations with Comments on the Heston Model

Stochastic differential equations have been an important tool in modeling complex financial relations, equipped with the possibility of being multidimensional to better oversee complexities inherent in finance. This multidimensionality, however, comes with a larger parameter space to estimate. Therefore, via a dimension reduction method, Sliced Inverse Regression, we aim to reduce high-dimensional parameter space to a reduced feature space and aim to estimate the parameters on this new featured space rather than using full data structure to lower computational costs. For this study, we closely study the Heston model, and remark our methodology of inference on this chosen model.

q-fin.MF↗

arXiv2025

Distributional Reinforcement Learning on Path-dependent Options

We reinterpret and propose a framework for pricing path-dependent financial derivatives by estimating the full distribution of payoffs using Distributional Reinforcement Learning (DistRL). Unlike traditional methods that focus on expected option value, our approach models the entire conditional distribution of payoffs, allowing for risk-aware pricing, tail-risk estimation, and enhanced uncertainty quantification. We demonstrate the efficacy of this method on Asian options, using quantile-based value function approximators.

q-fin.MF↗

arXiv2023

The QLBS Model within the presence of feedback loops through the impacts of a large trader

We extend the QLBS model by reformulating via considering a large trader whose transactions leave a permanent impact on the evolution of the exchange rate process and therefore affect the price of contingent claims on such processes. Through a hypothetical limit order book we quantify the exchange rate altered by such transactions. We therefore define the quoted exchange rate process, for which we assume the existence of a postulated hedging strategy. Given the quoted exchange rate and postulated hedging strategy, we find an optimal hedging strategy through batch-mode reinforcement learning given the trader alters the course of the exchange rate process. We assume that the trader has its own concept of fair price and we define our problem as finding the hedging strategy with much lower transaction costs yet delivering a price that well converges to the fair price of the trader. We show our contribution results in an optimal hedging strategy with much lower transaction costs and convergence to the fair price is obtained assuming sensible parameters.

q-fin.MF↗