Searcharxiv⌕ Search

arXiv · 2611.10574

Is price-impact decay statistically identifiable by propagator estimation? Diffusive prices force the inferred kernel to be transient and fine-tuned

Abstract

Despite the long memory of market-order flow, price dynamics remain diffusive---a seemingly paradoxical coexistence known in econophysics as the diffusivity puzzle. The propagator model has long been regarded as a promising solution: price impact decays in time, finely balanced against the order-flow memory, and propagators inferred from data have been reported to be consistent with this fine balance. Here we prove a no-go theorem: asymptotic impact decay is not statistically identifiable by propagator estimation, since for diffusive prices the fine-tuned decay is returned whether or not the true impact decays. It rests on universal theorems for the Wiener--Hopf deconvolution by which propagators are inferred under long-memory order flow: (i) the inferred propagator is necessarily transient, even when the true impact is permanent; (ii) under a mild condition, it decays with the fine-balance exponent, fixed by the order-flow autocorrelation alone. The fine-tuned decay is thus a property of the deconvolution, not evidence of impact decay. We demonstrate this with an exactly solvable null model that retains order splitting and the square-root law but whose impact is strictly permanent: the inferred propagator shows an artificial fine-tuned decay although the true impact never decays. Nonlinear propagator estimation, being of Wiener--Hopf type, is subject to the same theorems whenever true metaorders have long memory.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Shunta Fujiwara, Yuki Sato, Kiyoshi Kanazawa. 2026-11-07. Is price-impact decay statistically identifiable by propagator estimation? Diffusive prices force the inferred kernel to be transient and fine-tuned. https://searcharxiv.com/abs/2611.10574

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related papers

Measuring price impact and information content of trades in a time-varying setting

We propose a non-linear observation-driven version of the Hasbrouck vector auto-regression framework for dynamically estimating trades' market impact and information content. We find that market impact displays an intraday pattern superimposed with large fluctuations. Some of them are exogenous, and, as an example, we investigate market impact dynamics around FOMC announcements. Unlike the vector autoregression results, we reveal that the information content of the trades depends on the local liquidity level and the recent history of prices and trades. Finally, we use the model to estimate the time-varying permanent impact parameter, which allows performing a dynamic transaction cost analysis.

q-fin.TR↗

SAiFE-gym: Model-based Environments for Automated Market Making with Concentrated Liquidity

We present SAiFE_gym, a Python module that provides a collection of simulation environments for studying trading problems in constant product markets (CPMs) with concentrated liquidity (CL). These markets give liquidity providers (LPs) granular control over how their capital is allocated and enable them to adjust their range of liquidity provision dynamically based on market conditions, which in turn determines how they earn fees. We decompose the microstructure of CPMs with CL into interactive components that allow researchers and practitioners to capture various economic settings. We employ a vectorized approach to optimize our environments, making them scalable for high-dimensional reinforcement learning (RL) workflows, suitable for learning dynamic liquidity provision strategies. We demonstrate the benefits of our environments by evaluating the performance of RL agents in CPMs with CL under various market conditions.

q-fin.TR↗

Who Should Be on the Panel? Task-Conditioned Persona Selection for LLM Trading Agents

Multi-agent LLM systems are built by fixing a cast of roles and then designing how the cast coordinates. Which agents should make up the group for a given task is left to the designer, and the same cast is reused for every task. Financial trading makes the cost of that choice visible: stocks trade differently, and the trading styles suited to one need not suit another. We study task-conditioned persona selection, composing the group from the characteristics of the task it will face, and instantiate it with FinPanel, an LLM multi-agent trading system in which a recruiter selects a panel of named trader personas for each stock from a direction-blind statistical sketch of how the stock trades, each persona gives an independent daily call in its own style, and a manager commits the trade after reading the panel's arguments and their desk-kept, hierarchical track records. Because nothing is returned to the traders, the panel stays the panel that was selected, and it can trade a new stock from an empty record. On eight US stocks over the out-of-knowledge year 2025, against three prior LLM trading agents re-implemented in the same framework and against rule-based, predictive and reinforcement-learning baselines, the selected panel is ahead of a fixed panel of the same size and of the prior agents run as their papers prescribe, with a shallower drawdown than buy-and-hold on every stock; it keeps its performance with no warm-up and with no record at all, whereas the memory-augmented agents' returns swing with their warm-up; and neither more seats nor the manager's reading of the arguments substitutes for selecting the right ones. Who is in the group is a design decision to be made per task, before how the group coordinates.

q-fin.TR↗