arXiv paper shows optimal liquidation strategies are Lipschitz, enabling neural-net approximation
chaumian · x · 2026-11-13
A new arXiv paper (2611.14245) by Donnelly, Galimberti, Healy and Zhang studies the solution map of an optimal execution problem under general transient price impact. Building on Abi Jaber and Neuman's optimal strategy, the authors prove the strategy is locally Lipschitz in the model parameters, justifying neural-net approximations that take parameters as inputs. They propose methods to sample transient impact kernels for training data, project kernels to finite-dimensional neural inputs, and reduce dimensionality via the inventory–signal relationship, then use the trained net to quantify the financial cost of misspecifying the impact kernel.
More from Research
- TLAPS-proof LLM serving engine: 516,973 obligations at just 0.18% latency overhead — tianyin_xu · 2026-11-13
- Single-constraint KL minimization reduces to 1-D bisection on the dual — caglar_ee · 2026-11-13
- AI-Simulated Survey Matches Real Human Rankings: Spearman r=0.886 in Head-to-Head Match-ups — soumitrashukla9 · 2026-11-13
- Professor Uses AI Simulation to Pick His Executive Headshot from Six Looks, Biker Jacket Included — soumitrashukla9 · 2026-11-13
- Researcher pushes back on METR's 2x AI acceleration claim, citing 1.4x figure — HaydnBelfield · 2026-11-13
- Mercor launches OWE-Bench: 26 real engineering projects that stump coding agents — ben_burtenshaw · 2026-11-13