Paper: arXiv 2307.15805

Authors: Joffrey Derchu, Dimitrios Kavvathas, Thibaut Mastrolia, Mathieu Rosenbaum

Abstract

We study a toy two-player game for periodic double auction markets to generate liquidity. The game has imperfect information, which allows us to link market spreads with signal strength. We characterize Nash equilibria in cases with or without incentives from the exchange. This enables us to derive new insights about price formation and incentives design. We show in particular that without any incentives, the market is inefficient and does not lead to any trade between market participants. We however prove that quadratic fees indexed on each players half spread leads to a transaction and we propose a quantitative value for the optimal fees that the exchange has to propose in this model to generate liquidity.

Complexity vs Empirical Score

  • Math Complexity: 8.0/10
  • Empirical Rigor: 3.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper introduces a game-theoretic model with Gaussian signals and derives Nash equilibria, involving advanced probability and statistics. However, it is purely theoretical with no backtesting, datasets, or code, focusing solely on analytical proofs.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Generate liquidity in<br>periodic double auctions"] --> B["Methodology: Toy 2-Player Game<br>Imperfect Information &<br>Nash Equilibrium Analysis"]
  B --> C["Simulation Inputs:<br>Signal Strength &<br>Spread Parameters"]
  C --> D{"Analysis: Nash Equilibrium<br>without incentives?"}
  D -- No --> E["Outcome: Market Inefficiency<br>No Trade Occurs"]
  D -- Yes --> F["Outcome: Trade Possible"]
  F --> G["Key Finding:<br>Quadratic fees on half-spread<br>generate liquidity"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes quadratic fees indexed on each player's half‑spread to generate liquidity and derives a quantitative optimal fee value for the exchange.

Source sentence
We however prove that quadratic fees indexed on each players half spread leads to a transaction and we propose a quantitative value for the optimal fees that the exchange has to propose in this model to generate liquidity.

Assumes 6

  • The market is illiquid, with low trading frequency, small volumes, limited data and scarce information. market
    Source sentence
    By illiquid we mean a market where trading frequency is low, volumes are small, market data is limited and information is scarce.
  • The analysis uses a one‑period framework with exactly two participants (one buyer and one seller) taking part in an auction. model
    Source sentence
    We model this market by a very simplified one period framework with two participants, one buyer and one seller, taking part to an auction organized by an exchange.
  • The increment between the fair price and the last traded price is a centered Gaussian variable with variance v². statistical
    Source sentence
    At the time of the auction, we assume that the increment between the fair price, denoted by P∞and the last traded price is given by a centered Gaussian variable with variance v2.
  • Each participant has a proxy of the fair price (P∞|a for the seller, P∞|b for the buyer) with error standard deviations σa and σb drawn from independent random variables. model
    Source sentence
    We assume both participants have access to a proxy of the fair price at the beginning of the auction, denoted by P∞|a for the seller and P∞|b for the buyer. The standard deviations of the error of this proxy are drawn from independent random variables, denoted by σa and σb.
Show 2 more
  • Both participants use the same pricing rule, i.e., the deviation function δ does not depend on the agent (symmetric case). model
    Source sentence
    More precisely, we are interested in the symmetric case where market participants have the same pricing rule and consequently the function δi does actually not depend on i.
  • Agents are risk‑neutral and choose limit order prices to maximize the expected difference between the mid‑price and the fair price, accounting for the probability of no trade. model
    Source sentence
    So, for a given function δ, each (risk neutral) agent optimizes simultaneously the price of its limit order by maximizing or minimizing the expected value of the difference between the obtained midprice and the fair price, taking into account the probability that no trade happens.

Data

No dataset named in the text.

Costs

Not applicable The paper reports no trading or execution result; the cost the extractor found is a model parameter.

The paper-text extraction found a cost parameter inside a model (Quadratic transaction fees are imposed on each player's half‑spread (parameter γ) to induce trade.); the classifier found no reported trading result, so costs are not applicable.

Finds 3

  • Without any incentives, the market is inefficient and no trade occurs between participants. negative
    Source sentence
    We show in particular that without any incentives, the market is inefficient and does not lead to any trade between market participants.
  • Introducing quadratic fees indexed on each player's half‑spread generates a transaction and creates liquidity. theoretical
    Source sentence
    We however prove that quadratic fees indexed on each players half spread leads to a transaction and we propose a quantitative value for the optimal fees that the exchange has to propose in this model to generate liquidity.
  • The optimal fee γ* is uniquely determined and given by γ* = Φ′(y*) / (2√2 σ+), where y* solves 1−Φ(y*)−y*Φ′(y*) = 0. theoretical
    Source sentence
    It is given by γ∗= Φ′(y∗) 2 √ 2σ+ where y∗is the unique solution to 1 −Φ(y∗) −y∗Φ′(y∗) = 0.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 10 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 3 papers in the archive

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.