arXiv subjects

Aaron Wheeler

Publications and source records attributed to Aaron Wheeler.

3 recordsLinked to original sources

arXiv2024

MarketGPT: Developing a Pre-trained transformer (GPT) for Modeling Financial Time Series

This work presents a generative pre-trained transformer (GPT) designed for modeling financial time series. The GPT functions as an order generation engine within a discrete event simulator, enabling realistic replication of limit order book dynamics. Our model leverages recent advancements in large language models to produce long sequences of order messages in a steaming manner. Our results demonstrate that the model successfully reproduces key features of order flow data, even when the initial order flow prompt is no longer present within the model's context window. Moreover, evaluations reveal that the model captures several statistical properties, or 'stylized facts', characteristic of real financial markets and broader macro-scale data distributions. Collectively, this work marks a significant step toward creating high-fidelity, interactive market simulations.

q-fin.TR↗

arXiv2023

Scalable Agent-Based Modeling for Complex Financial Market Simulations

In this study, we developed a computational framework for simulating large-scale agent-based financial markets. Our platform supports trading multiple simultaneous assets and leverages distributed computing to scale the number and complexity of simulated agents. Heterogeneous agents make decisions in parallel, and their orders are processed through a realistic, continuous double auction matching engine. We present a baseline model implementation and show that it captures several known statistical properties of real financial markets (i.e., stylized facts). Further, we demonstrate these results without fitting models to historical financial data. Thus, this framework could be used for direct applications such as human-in-the-loop machine learning or to explore theoretically exciting questions about market microstructure's role in forming the statistical regularities of real markets. To the best of our knowledge, this study is the first to implement multiple assets, parallel agent decision-making, a continuous double auction mechanism, and intelligent agent types in a scalable real-time environment.

q-fin.TR↗

arXiv2021

Structure-Property-Performance Relationships of Cuprous Oxide Nanostructures for Dielectric Mie Resonance-Enhanced Photocatalysis

Nanostructured metal oxides, such as Cu2O, CeO2, {\alpha}-Fe2O3, and TiO2 can efficiently mediate photocatalysis for solar-to-chemical energy conversion and pollution remediation. In this contribution, we report a novel approach, dielectric Mie resonance-enhanced photocatalysis, to enhance the catalytic activity of metal oxide photocatalysts. Specifically, we demonstrate that Cu2O nanostructures exhibiting dielectric Mie resonances can exhibit up to an order of magnitude higher photocatalytic rate as compared to Cu2O nanostructures not exhibiting dielectric Mie resonances. Our finite-difference time-domain (FDTD) simulation and experimental results predict a volcano-type relationship between the photocatalytic rate and the size of Cu2O nanospheres and nanocubes. Using transient absorption measurements, we reveal that a coherent electronic process associated with dielectric Mie resonance-mediated charge carrier generation is dominant in Cu2O nanostructures that exhibit higher photocatalytic rates. Although we experimentally demonstrate dielectric Mie resonance-enhanced photocatalysis using Cu2O particles here, based on our FDTD simulations, we anticipate the same can be achieved with other metal oxide photocatalysts, including CeO2, {\alpha}-Fe2O3, and TiO2.

physics.app-ph↗