Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1]
Welfare Effects of Market Making in Continuous Double Auctions
Authors: Elaine Wah, Mason Wright, Michael P. Wellman
JAIR 2017 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We employ empirical simulationbased methods to evaluate heuristic strategies for market makers as well as background investors in a variety of complex trading environments. Our market model incorporates private and common valuation elements, with dynamic fundamental value and asymmetric information. In this context, we compare the surplus achieved by background traders in strategic equilibrium, with and without a market maker. Our findings indicate that the presence of the market maker strongly tends to increase total welfare across various environments. |
| Researcher Affiliation | Collaboration | Elaine Wah EMAIL IEX Group, Inc., 4 World Trade Center 150 Greenwich St, 44th Floor, New York, NY 10007 Mason Wright EMAIL Michael P. Wellman EMAIL Computer Science & Engineering, University of Michigan 2260 Hayward St, Ann Arbor, MI 48109 |
| Pseudocode | No | The paper describes strategies for background traders (Section 4.2) and market makers (Section 4.3) in detail, including mathematical formulations, but it does not present these strategies in a structured pseudocode or algorithm block. The descriptions are in paragraph form. |
| Open Source Code | Yes | The source code of the financial market simulator employed for this study is publicly available online (Strategic Reasoning Group, 2016). |
| Open Datasets | No | The paper describes a simulation-based approach where data is generated within the market environment model rather than using pre-existing public datasets. For example, it states: 'We generate Θi from a set of 2qmax values drawn independently from a Gaussian distribution.' and 'We evaluate the performance of background traders and the MM within 30 parametrically distinct environments.' |
| Dataset Splits | No | The paper does not use traditional train/test/validation splits for a fixed dataset. Instead, it describes how simulation runs are managed within an Empirical Game-Theoretic Analysis (EGTA) framework: 'From extensive simulation over thousands of strategy profiles, we estimate game models for various instances of the target scenario.' and 'For each equilibrium, we estimated background-trader surplus by sampling 2,500 profiles according to the equilibrium mixture, running 25–100 simulations per sampled profile (at least 62,500 simulations in total) and then recording the aggregate surplus.' |
| Hardware Specification | No | We utilize the EGTAOnline infrastructure (Cassell & Wellman, 2013) for conducting and managing our experiments, and run our simulations on a high-performance computing cluster at the University of Michigan. |
| Software Dependencies | No | The paper mentions 'EGTAOnline infrastructure (Cassell & Wellman, 2013)' and 'financial market simulator... (Strategic Reasoning Group, 2016)' but does not provide specific version numbers for these or any other software components. |
| Experiment Setup | Yes | We evaluate the performance of background traders and the MM within 30 parametrically distinct environments. ... The global fundamental has a mean value r = 105 and mean-reversion parameter κ = 0.05. The minimum tick size pts is fixed at 1. The maximum number of units the background trader can be long or short at any time is qmax = 10. If present, the MM in each environment enters the market at the start of the simulation and reenters with rate λM M = 0.005, or approximately once every 200 time steps. ... Each MM strategy type employs a fixed spread ω {64, 128, 256, 512, 1024}. |