Steve Y. Yang

An Agent Based Model of the E-Mini S&P 500 and the Flash Crash

We propose a near zero-intelligence agent-based model of the E-Mini S&P 500 futures market that allows for a close examination of market microstructure in the context of a flash crash. Several classes of agents are characterized by how fast they trade and where they place trades in the limit order book. These agents’ orders populate the simulated market in a way consistent with real world participation rates. The simulated market is validated against important empirically observed characteristics of price returns and volatility. Additionally, to illustrate the applicability of the simulation experimental results are present, which examine the leading hypothesis for the cause of the May 6th 2010 Flash Crash.

Twitter financial community sentiment and its predictive relationship to stock market movement

Twitter, one of the several major social media platforms, has been identified as an influential factor for financial markets by multiple academic and professional publications in recent years. The motivation of this study hinges on the growing popularity of the use of Twitter and the increasing prevalence of its influence among the financial investment community. This paper presents empirical evidence of the existence of a financial community on Twitter in which users’ interests align with financial market-related topics. We establish a methodology to identify relevant Twitter users who form the financial community, and we also present the empirical findings of network characteristics of the financial community. We observe that this financial community behaves similarly to a small-world network, and we further identify groups of critical nodes and analyse their influence within the financial community based on several network centrality measures. Using a novel sentiment analysis algorithm, we construct a weighted sentiment measure using tweet messages from these critical nodes, and we discover that it is significantly correlated with the returns of the major financial market indices. By forming a financial community within the Twitter universe, we argue that the influential Twitter users within the financial community provide a proxy for the relationship between social sentiment and financial market movement. Hence, we conclude that the weighted sentiment constructed from these critical nodes within the financial community provides a more robust predictor of financial markets than the general social sentiment.

Agent based model of the e-mini future: application for policy making

An agent-based model (ABM) has a structure, which includes a set of agents, a topology and an environment. A simplified conception of a financial market includes a set of market participants, a trading mechanism, and a set of securities. In a typical ABM of a financial market, the market participants are agents, the market mechanism is the topology and the exogenous flow of information into the market is the environment. A zero-intelligence ABM model of the E-Mini Futures Market is presented. Several classes of agents are characterized by their speed and placement of orders within the limit order book. The proposed minimum quote life rule is implemented in the simulation. The minimum quote life rule prevents new orders from being cancelled or modified before a given time limit. Through experimentation, tradeoff curves are generated. Thereby, illustrating the usefulness of this ABM and its ability to inform ongoing financial policy debates.

Gaussian process-based algorithmic trading strategy identification

Many market participants now employ algorithmic trading, commonly defined as the use of computer algorithms, to automatically make certain trading decisions, submit orders and manage those orders after submission. Identifying and understanding the impact of algorithmic trading on financial markets has become a critical issue for market operators and regulators. Advanced data feeds and audit trail information from market operators now allow for the full observation of market participants’ actions. A key question is the extent to which it is possible to understand and characterize the behaviour of individual participants from observations of trading actions. In this paper, we consider the basic problems of categorizing and recognizing traders (or, equivalently, trading algorithms) on the basis of observed limit orders. These problems are of interest to regulators engaged in strategy identification for the purposes of fraud detection and policy development. Methods have been suggested in the literature for describing trader behaviour using classification rules defined over a feature space consisting of summary trading statistics of volume and inventory, along with derived variables that reflect the consistency of buying or selling behaviour. Our principal contribution is to suggest an entirely different feature space that is constructed by inferring key parameters of a sequential optimization model that we take as a surrogate for the decision-making process of the traders. In particular, we model trader behaviour in terms of a Markov decision process. We infer the reward (or objective) function for this process from observations of trading actions using a process from machine learning known as inverse reinforcement learning (IRL). The reward functions learned through IRL then constitute a feature space that can be the basis for supervised learning (for classification or recognition of traders) or unsupervised learning (for categorization of traders). Making use of a real-world data-set from the E-Mini futures contract, we compare two principal IRL variants, linear IRL and Gaussian Process IRL, against a method based on summary trading statistics. Results suggest that IRL-based feature spaces support accurate classification and meaningful clustering. Further, we argue that, because they attempt to learn traders’ underlying value propositions under different market conditions, the IRL methods are more informative and robust than the summary statistic-based approach and are well suited for discovering new behaviour patterns of market participants.

Twitter financial community modeling using agent based simulation

With the empirical evidence that Twitter influences the financial market, there is a need for a bottom-up approach focusing on individual Twitter users and their message propagation among a selected Twitter community with regard to the financial market. This paper presents an agent-based simulation framework to model the Twitter network growth and message propagation mechanism in the Twitter financial community. Using the data collected through the Twitter API, the model generates a dynamic community network with message propagation rates by different agent types. The model successfully validates against the empirical characteristics of the Twitter financial community in terms of network demographics and aggregated message propagation pattern. Simulation of the 2013 Associated Press hoax incident demonstrates that removing critical nodes of the network (users with top centrality) dampens the message propagation process linearly and critical node of the highest betweenness centrality has the optimal effect in reducing the spread of the malicious message to lesser ratio of the community.

Stock portfolio selection using learning-to-rank algorithms with news sentiment

Abstract In this study, we apply learning-to-rank algorithms to design trading strategies using relative performance of a group of stocks based on investors’ sentiment toward these stocks. We show that learning-to-rank algorithms are effective in producing reliable rankings of the best and the worst performing stocks based on investors’ sentiment. More specifically, we use the sentiment shock and trend indicators introduced in the previous studies, and we design stock selection rules of holding long positions of the top 25% stocks and short positions of the bottom 25% stocks according to rankings produced by learning-to-rank algorithms. We then apply two learning-to-rank algorithms, ListNet and RankNet, in stock selection processes and test long-only and long-short portfolio selection strategies using 10 years of market and news sentiment data. Through backtesting of these strategies from 2006 to 2014, we demonstrate that our portfolio strategies produce risk-adjusted returns superior to the S&P 500 index return, the hedge fund industry average performance - HFRIEMN, and some sentiment-based approaches without learning-to-rank algorithm during the same period.

An adaptive portfolio trading system: A risk-return portfolio optimization using recurrent reinforcement learning with expected maximum drawdown

A reinforcement learning trading algorithm with expected drawdown risk is proposed.The expected maximum drawdown is shown to improve portfolio signal generation.The effectiveness of the method is validated using different transaction costs.An adaptive portfolio rebalancing system with automated retraining is recommended. Dynamic control theory has long been used in solving optimal asset allocation problems, and a number of trading decision systems based on reinforcement learning methods have been applied in asset allocation and portfolio rebalancing. In this paper, we extend the existing work in recurrent reinforcement learning (RRL) and build an optimal variable weight portfolio allocation under a coherent downside risk measure, the expected maximum drawdown, E(MDD). In particular, we propose a recurrent reinforcement learning method, with a coherent risk adjusted performance objective function, the Calmar ratio, to obtain both buy and sell signals and asset allocation weights. Using a portfolio consisting of the most frequently traded exchange-traded funds, we show that the expected maximum drawdown risk based objective function yields superior return performance compared to previously proposed RRL objective functions (i.e. the Sharpe ratio and the Sterling ratio), and that variable weight RRL long/short portfolios outperform equal weight RRL long/short portfolios under different transaction cost scenarios. We further propose an adaptive E(MDD) risk based RRL portfolio rebalancing decision system with a transaction cost and market condition stop-loss retraining mechanism, and we show that the proposed portfolio trading system responds to transaction cost effects better and outperforms hedge fund benchmarks consistently.

News Sentiment to Market Impact and its Feedback Effect

Although market feedback on investor sentiment effect has been conceptually identified in the existing finance literature and investment strategies have been designed to explore this effect, there lacks systematic analysis in a quantified manner on such effect. Digitization of news articles and the advancement of computational intelligence applications have led to a growing influence of news sentiment over financial markets in recent years. News sentiment has often been used as a proxy for gauging investor sentiment and reflecting the aggregate confidence of the society toward future market. Previous studies have primarily focused on elucidating the unidirectional impact of news sentiment on market returns and not vice versa. In this study, we analyze more than 12 millions of news articles and document the presence of a significant feedback effect between news sentiment and market returns across the major indices in the US financial market. More specifically, we find that news sentiment exhibits a lag-5 effect on market returns and conversely market returns elicit consistent lag-1 effects on news sentiment. This aligns well with our intuition that news sentiment drives trading activity and investment decisions. In turn, heightened investment activity further stimulates involuntary responses, which manifest in the form of more news coverage and publications. The evidence presented highlights the strong correlation between news sentiment and market returns and demonstrates the benefits of advancing knowledge in data-driven modeling and its interaction with market movements.

A study of dark pool trading using an agent-based model

A dark pool is a securities trading venue with no published market depth feed. Such markets have traditionally been utilized by large institutions as an alternative to public exchanges to execute large block orders which might otherwise impact settlement price. It is estimated that the trading volume of dark pool markets was 9% to 12% of the total U.S. equity market share volume in 2010 [1]. This phenomenon raises questions regarding the fundamental value of securities traded through dark pool markets and their impact on the price discovery process in traditional “visible” markets. In this paper, we establish a modeling framework for dark pool markets through agent-based modeling. It presents and validates the costs and benefits of trading small orders in dark pool markets. Simulated trading of 78 selected stocks demonstrates that dark pool market traders can obtain better execution rate when the dark pool market has more uninformed traders relative to informed traders. In addition, trading stocks with larger market capitalization yields better price improvement in dark pool markets.

Agent Based Model of the E-Mini S&P 500 Future: Application for Policy Making

An agent-based model (ABM) has a structure, which includes a set of agents, a topology and an environment. A simplified conception of a financial market includes a set of market participants, a trading mechanism, and a set of securities. In a typical ABM of a financial market, the market participants are agents, the market mechanism is the topology and the exogenous flow of information into the market is the environment. A zero-intelligence ABM model of the E-Mini Futures Market is presented. Several classes of agents are characterized by their speed and placement of orders within the limit order book. The proposed minimum quote life rule is implemented in the simulation. The minimum quote life rule prevents new orders from being cancelled or modified before a given time limit. Through experimentation, trade-off curves are generated. Thereby, illustrating the usefulness of this ABM and its ability to inform ongoing financial policy debates.