Market Making with Fads, Informed, and Uninformed Traders

Emilio Barucci, Adrien Mathieu, Leandro Sánchez-Betancourt · 2025

Read Market Making with Fads, Informed, and Uninformed Traders by Emilio Barucci, Adrien Mathieu, Leandro Sánchez-Betancourt (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-03658.txt.

Introduction

Emilio Barucci, Adrien Mathieu, and Leandro Sanchez-Betancourt model a market maker who faces fad-driven prices in 2025. Prices drift from true value in short bursts, and both informed and uninformed traders arrive at rates that shift with the fad. The maker sees orders but not who sent them or the fad state.

Both full and partial sight versions of the control task are solved in the paper. The team prices the worth of sight and the cost of liquidity as the share of informed traders shifts. Fit steps from market data and gains over fad-blind plans close the work.

Catalog record

CREATOREmilio Barucci, Adrien Mathieu, Leandro Sánchez-Betancourt
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE137 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-03658

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Market Making with Fads, Informed, and Uninformed Traders Emilio Barucci, Adrien Mathieu, Leandro Sánchez-Betancourt Abstract We characterise the solution to a continuous-time optimal liquidity provision problem in a market populated by informed and uninformed traders. In our model, the asset price exhibits fads —these are short-term deviations from the fundamental value of the asset. Conditional on the value of the fad, we model how informed traders and uninformed traders arrive in the market. The market maker is aware of the existence of the two groups of traders but only observes the anonymous order arrivals. We study both, the complete and the partial information versions of the control problem faced by the market maker. In such frameworks, we characterise the value of information, and we find the price of liquidity as a function of the proportion of informed traders in the market. We discuss how to calibrate model parameters from market data and we quantify the outperformance over strategies that ignore these unobservable short-term deviations. † † footnotetext: Email addresses: [email protected] (Adrien Mathieu), [email protected] (Leandro Sánchez-Betancourt), [email protected] (Emilio Barucci). † † footnotetext: Corresponding author details: Leandro Sánchez-Betancourt, Mathematical Institute, University of Oxford, Radcliffe Observatory, Andrew Wiles Building, Woodstock Rd, Oxford OX2 6GG. † † footnotetext: LBS and AM would like to acknowledge support from the Oxford-Man Institute of Quantitative Finance. † † footnotetext: The data that support the findings of this study are available from the corresponding author upon reasonable request. † † footnotetext: The authors declare that they have no conflicts of interest to report. Lastly, for the partial information setup, we explore how to go beyond the Kalman-Bucy filter to extract information about the fad from the market arrivals. Keywords: market making, signals, informed traders, noise trading, stochastic filtering, fad 1 Introduction The market making problem has been studied in a large body of literature dealing mainly with two issues: asymmetric information and inventory management; see the seminal contribution of [ 29 ] and [ 2 , 34 ] , respectively. 1 1 1 See also the books [ 27 , 33 , 14 ] for a comprehensive account. Asymmetric information is concerned with the financial losses from trading with agents who have privileged information about the fundamental value of the asset (adverse selection). Inventory management focuses on controlling the risk associated with an open inventory, that is, the risk that the price goes up when the inventory is negative (short position) or that the price goes down when the inventory is positive (long position). In what follows, we approach the problem through an angle that combines asymmetric information and inventory management in a new way: we study the market making problem when the market is affected by “fads”. By fads we mean short-term deviations from the fundamental price —a component that is associated with market inefficiencies and that may be related to over-optimism, short-termism, or other behavioural explanations (e.g. herding); see [ 28 , 39 ] . The fad affects the market in two different ways. First, it affects the price of the asset. In addition to the classical arrival of information, which is modelled as a Brownian motion, there is a mean reverting process (the fad) that drives mid-prices —the mean-reversion well describes the fad’s long run bursting properties. Second, the fad also affects the arrival of market orders. This is because a fraction of agents populating the market (informed traders) distinguish fads from fundamental values. The objective of the market maker is to manage inventory while facing both (i) uninformed traders, who are sensitive to the displacements of bid-ask quotes from the mid-price as in [ 2 ] , and (ii) informed traders, who identify the true fundamental value of the asset (disentangling it from the fad component) and arrive in the market according to the displacements of the quotes from the fundamental. The problem crucially depends on the information of the market maker about the fad. In this paper, we work out optimal bid-ask pricing strategies under both, a perfect information setup and an incomplete information setup. This allows us to obtain insights into the value of information depending on the proportion of informed traders in the market. We find that the optimal displacements are affected by the fad (or its filter) asymmetrically: as the fad increases, the market maker decreases the price of liquidity in the ask side and increases the price of liquidity in the bid side to manage the inventory optimally. This arises because the market maker aims to balance the arrival of trades from informed traders (who consider the fundamental value rather than the mid-price as reference price) and uninformed traders (who consider the mid-price as reference price). This asymmetric effect represents a novelty of our model over the classical Avellaneda-Stoikov framework [ 2 ] : private information (or its filter) plays a key role in the quote asymmetry we characterise; this is in addition to the classical asymmetry coming from inventory management. Confirming this interpretation, in our analysis, we show that when the trades sent by informed traders become “sharper” or more “toxic”, i.e., trades carried out when the fundamental value of the asset is farther from the mid-price, the market maker’s performance worsens and the bid-ask spread increases. Our paper is closely related to the strand of the literature that studies how brokers deal with toxic order flow. Toxic flow is the trading activity of informed traders who have access to signals about the short-term trend of the asset and exploit them to their advantage. For example, [ 16 ] analyses a model where informed and uninformed traders carry out their trading with a broker. 2 2 2 See [ 13 , 23 ] for methods to detect toxic order flow. In their model, private information is described by a stochastic process that affects the fundamental value of the asset. Similarly, in our model, the market price is affected by a mean reverting process describing deviations (fad) from the fundamental. Conditional on the fad, we model how informed traders and uninformed traders arrive in the market in a framework similar to [ 2 ] . 3 3 3 An interesting example of a mean-field game of strategic players interacting in an Avellaneda-Stoikov framework is [ 4 ] . To the best of our knowledge, this is the first paper where the order arrival from informed and uninformed traders is incorporated in the Avellaneda-Stoikov framework. Since the reference paper by [ 1 ] , several papers have incorporated predictive signals into a stochastic control framework of execution cost minimisation. Predictive signals refer mainly to measurable processes built from order book dynamics, e.g., the order flow or the order book imbalance, which affects the asset price in the short term; see, e.g., [ 21 , 15 ] . There are a number of contributions in the optimal execution literature that address the problem in a full information setting, see e.g., [ 15 , 37 , 38 , 5 ] , other papers that consider a noisy signal of future stock prices, see e.g., [ 6 ] , and others that deal with the incomplete information setting, see e.g., [ 18 ] . Within the market making problem there are fewer examples.

Read the complete text