Artificial Market Simulation Research Articles

Analysis of Optimal Market Making Strategies of CFD Market Using Artificial Market Simulation

Analysis of Optimal Market Making Strategies of CFD Market Using Artificial Market Simulation

PAMS: Platform for Artificial Market Simulations

PAMS: Platform for Artificial Market Simulations

Impact Analysis of Financial Regulation on Multi-Asset Markets Using Artificial Market Simulations

In this study, we assessed the impact of capital adequacy ratio (CAR) regulation in the Basel regulatory framework. This regulation was established to make the banking network robust. However, a previous work argued that CAR regulation has a destabilization effect on financial markets. To assess impacts such as destabilizing effects, we conducted simulations of an artificial market, one of the computer simulations imitating real financial markets. In the simulation, we proposed and used a new model with continuous double auction markets, stylized trading agents, and two kinds of portfolio trading agents. Both portfolio trading agents had trading strategies incorporating Markowitz’s portfolio optimization. Additionally, one type of portfolio trading agent was under regulation. From the simulations, we found that portfolio optimization as each trader’s strategy stabilizes markets, and CAR regulation destabilizes markets in various aspects. These results show that CAR regulation can have negative effects on asset markets. As future work, we should confirm these effects empirically and consider how to balance between both positive and negative aspects of CAR regulation.

Deep Reinforcement Learning in Agent Based Financial Market Simulation

Prediction of financial market data with deep learning models has achieved some level of recent success. However, historical financial data suffer from an unknowable state space, limited observations, and the inability to model the impact of your own actions on the market can often be prohibitive when trying to find investment strategies using deep reinforcement learning. One way to overcome these limitations is to augment real market data with agent based artificial market simulation. Artificial market simulations designed to reproduce realistic market features may be used to create unobserved market states, to model the impact of your own investment actions on the market itself, and train models with as much data as necessary. In this study we propose a framework for training deep reinforcement learning models in agent based artificial price-order-book simulations that yield non-trivial policies under diverse conditions with market impact. Our simulations confirm that the proposed deep reinforcement learning model with unique task-specific reward function was able to learn a robust investment strategy with an attractive risk-return profile.

Does an Artificial Intelligence Perform Market Manipulation With Its Own Discretion? – A Genetic Algorithm Learns in an Artificial Market Simulation -

Who should be charged with responsibility for an artificial intelligence performing market manipulation have been discussed. In this study, I constructed an artificial intelligence using a genetic algorithm that learns in an artificial market simulation, and investigated whether the artificial intelligence discovers market manipulation through learning with an artificial market simulation despite a builder of artificial intelligence has no intention of market manipulation. As a result, the artificial intelligence discovered market manipulation as an optimal investment strategy. This result suggests necessity of regulation, such as obligating builders of artificial intelligence to prevent artificial intelligence from performing market manipulation.

Effects Analysis of CAR Regulations on Financial Markets using Artificial Market Simulations

Effects Analysis of CAR Regulations on Financial Markets using Artificial Market Simulations

A Modelling of Bitcoin Price Dynamics by the Artificial Market Algorithm

A famous crypto-currency “Bitcoin” was invented in 2009. Since then, we have observed dramatic fluctuation of its price due to the nature of crypto-currency. In other words, the level of the fluctuation seems totally different from the ordinary financial assets like equities, interest rates and foreign exchange rates. Therefore it is afraid that the well-known framework of PDE used for those financial assets, is not suitable to express Bitcoin’s dynamics. In this article, an algorithm of artificial market simulation model, so called “150 Agents Model” is proposed to express the fluctuation of Bitcoin price. The model is composed of two types of agents, Traders and Non-Traders, who trade Bitcoin through “itayose” method. The aim of the model is not pricing Bitcoin exactly for trading. Rather it could visualize the structure implied behind the price actions. By calibrating the model to actual price fluctuation of Bitcoin, we are able to see the overall market picture composed of agents to recognize their target prices and price sensitivities as the driver of the price dynamics, to predict the range of next big fluctuation and to recognize how the change of agent’s target price and its sensitivity to the price change, affect the landscape of the price range. This model is versatile and flexible. We could add new agents, take new issued amount into consideration, express flash crush and also simulate fluctuation of ordinary financial assets like foreign exchange rates by applying a moderate range of the parameters, which could produce a kind of Elliot Wave dynamics.

Platform design for large-scale artificial market simulation and preliminary evaluation on the K computer

Artificial market simulations have the potential to be a strong tool for studying rapid and large market fluctuations and designing financial regulations. High-frequency traders, that exchange multiple assets simultaneously within a millisecond, are said to be a cause of rapid and large market fluctuations. For such a large-scale problem, this paper proposes a software or computing platform for large-scale and high-frequency artificial market simulations (Plham: /pl $$\Lambda$$ m). The computing platform, Plham, enables modeling financial markets composed of various brands of assets and a large number of agents trading on a short timescale. The design feature of Plham is the separation of artificial market models (simulation models) from their execution (execution models). This allows users to define their simulation models without parallel computing expertise and to choose one of the execution models they need. This computing platform provides a prototype execution model for parallel simulations, which exploits the variety in trading frequency among traders, that is, the fact that some traders do not require up-to-date information of markets changing in millisecond order. We evaluated a prototype implementation on the K computer using up to 256 computing nodes.

Investigation of the rule for investment diversification at the time of a market crash using an artificial market simulation

As financial products have grown in complexity and level of risk compounding in recent years, investors have come to find it difficult to assess investment risk. Furthermore, companies managing mutual funds are increasingly expected to perform risk control and thus prevent assumption of unforeseen risk by investors. A related revision to the mutual fund legal system in Japan led to establishing what is known as “the rule for investment diversification” in December 2014, without a clear discussion of its expected effects on market price formation having taken place. In this paper, we therefore, used an artificial market to investigate its effects on price formation in financial markets where investors must follow the rule at the time of a market crash that was caused by the collapse of the asset fundamental price. As results, we found that, in a two-asset market where investors had to follow the rule for investment diversification, when the fundamental price of one asset collapsed and its market price also collapsed, the other asset market price also fell.

Micro-Foundation of ARCH Model

The Japanese version of this paper can be found at http://ssrn.com/abstract=2710516.Many macroeconomic study argued macroeconomic models should be aggregated by micro processes models (micro-foundation) and many micro-founded macroeconomic models were built. On the other hand, there are many models for price variation of a risk asset, which is macro phenomena, however, there are few studies for micro-foundation of such models. In this study we tried micro-foundation of an ARCH model using intelligence of artificial market simulation studies. That is we tried to clarify which micro processes determine each coefficient of an ARCH model. Then, we showed that the dispersion of investors' estimated prices is larger or the orders by the buy-sell imbalance taking liquidity are more, the volatility is larger. And we showed that the ration of the normal investors taking liquidity to the noise traders providing liquidity is higher or the measure of risk aversion of the normal investors is lower, the magnitude of volatility clustering is larger.

ARCHHHHHHHHHHHHHHHH (Micro-Foundation of ARCH Model)

The English version of this paper can be found at http://ssrn.com/abstract=2710457.Japanese Abstract: 経済のマクロモデルは，ミクロプロセスの積み上げによって構築する，いわいる``ミクロ的基礎付け''(micro-foundation)がなされるべきとの主張が多くなされ，ミクロ的基礎付けがなされた経済のマクロモデルが多く生まれた．一方，リスク資産の価格変動というマクロ量をモデル化したものは多く存在するが，これらのモデルをミクロ的基礎付けしようという試みは多くない．そこで本研究では人工市場研究からの知見を用いて，ARCHモデルのミクロ的基礎付け，すなわち，各係数がどのミクロプロセスから生じているのかを明らかにすることを試みた．その結果，投資家の予想価格のばらつきと需給の歪みが大きくなるとボラティリティは大きくなり，流動性を奪う投資家の存在割合が大きくなったり投資家のリスク回避度が小さくなるとボラティリティクラスタリングは大きくなることが分かった．English Abstract: Many macroeconomic study argued macroeconomic models should be aggregated by micro processes models (micro-foundation'') and many micro-founded macroeconomic models were built. On the other hand, there are many models for price variation of a risk asset, which is macro phenomena, however, there are few studies for micro-foundation of such models. In this study we tried micro-foundation of an ARCH model using intelligence of artificial market simulation studies. That is we tried to clarify which micro processes determine each coefficient of an ARCH model. Then, we showed that the dispersion of investors' estimated prices is larger or the orders by the buy-sell imbalance taking liquidity are more, the volatility is larger. And we showed that the ration of the normal investors taking liquidity to the noise traders providing liquidity is higher or the measure of risk aversion of the normal investors is lower, the magnitude of volatility clustering is larger.

A Review of Recent Artificial Market Simulation Studies for Financial Market Regulations And/Or Rules

A Review of Recent Artificial Market Simulation Studies for Financial Market Regulations And/Or Rules

Effects of Price Regulations and Dark Pools on Financial Market Stability: An Investigation by Multiagent Simulations

SummaryWe built an artificial market model and investigated the impact of large erroneous orders on financial market price formations. Comparing the case of consented large erroneous orders in the short term with that of continuous small erroneous orders in the long term, if amounts of orders are the same, we found that the orders induced almost the same price fall range. We also analysed effects of price variation limits for erroneous orders and found that price variation limits that employ a limitation term shorter than the time erroneous orders exist effectively prevent large price fluctuations. We also investigated effects of up‐tick rules, adopting the trigger method that the Japan Financial Services Agency adopted in November 2013.We also investigated whether dark pools that never provide any order books stabilize markets or not using the model including one lit market, which provides all order books to investors, and one dark pool. We found that markets become more stable as the dark pool is increasingly used. We also found that using the dark pool more reduces the market impacts. However, if other investors’ usage rates of dark pools become too large, investors must use the dark pool more than other investors to avoid market impacts. When a tick size of a lit market is larger, dark pools are more useful to avoid market impacts. These results suggest that dark pools stabilize markets when the usage rate is under some threshold and negatively affect the market when the usage rate is over that threshold. Our simulation results suggest the threshold might be much larger than the usage rate in present real financial markets.This study is the first to discuss whether or not several concrete and actually adoptable regulations, including those that have never been employed (e.g. price variation limits with various parameters), could prevent large fluctuations of market prices, including those beyond our experience, using artificial market simulations, and to discuss quantitatively how spreading of dark pools beyond our experience could affect market price formations using the artificial market simulations. In short, this study is the first study to comprehensively discuss how regulations and financial systems beyond our experience could affect price formations using the artificial market simulations. Copyright © 2015 John Wiley & Sons, Ltd.

Investigation of Price Variation Limits, Short Selling Regulation, and Uptick Rules and Their Optimal Design by Artificial Market Simulations

SUMMARYWe built an artificial market model and compared the effects of price variation limits, short sell regulations, and uptick rules. When there were no regulations, the price fell below the fundamental value when economic bubble collapses occurred. When there were regulations, this overshoot did not occur and the market was more effective. However, the short sell regulations and uptick rules caused the trading prices to be higher than the base value. To summarize these points, the price variation limits have the potential to make the market more effective. We also surveyed an adequate limitation price range and an adequate limitation time span for the price variation limit and found conditions for the parameters for the price variation limit that prevent such overshoots.

Investigation and Optimal Designing of a Price Variation Limit, a Short Sell Regulation and an Up-Tick Rule using by Artificial Market Simulations

Investigation and Optimal Designing of a Price Variation Limit, a Short Sell Regulation and an Up-Tick Rule using by Artificial Market Simulations

Analysis of the Impact of Changing Strategy on the GARCH Effect with Artificial Market Simulation

Analysis of the Impact of Changing Strategy on the GARCH Effect with Artificial Market Simulation

Design of Financial Market Regulations against Large Price Fluctuations Using by Artificial Market Simulations

We built an artificial market model and compared effects of price variation limits, short selling regulations and up-tick rules. In the case without the regulations, the price fell to below a fundamental value when an economic crush occurred. On the other hand, in the case with the regulations, this overshooting did not occur. However, the short selling regulation and the up-tick rule caused the trading prices to be higher than the fundamental value. We also surveyed an adequate limitation price range and an adequate limitation time span for the price variation limit and found a parameters’ condition of the price variation limit to prevent the over-shorts. We also showed the limitation price range should be bigger than a volatility calculated by the limitation time span.

Influence of Investors’ Loss-cut Behavior to Artificial Market

The aim of this study is to discuss the influence of investors’ loss-cut behavior to the market stock price and the investors’ asset distribution by using the artificial market simulation. The loss-cut behavior is strongly related to the anchoring effect, which is one of the investors’ cognitive biases. The anchoring effect is a psychological heuristic that influences the way people intuitively assess probabilities. The investor in the actual market is modeled as the agent in the artificial market. The agent has the prediction rule of the stock price and the anchoring coefficient related to the strength of the anchoring effect. Once the agent predicts the stock price by the prediction rule, he determines the order price and the order amount from the predicted stock price according to the strength of the anchoring effect. Agents’ orders are gathered to the market and the stock price is determined according to the Itayose method. In the numerical examples, two kinds of agents with different anchoring effects are modeled; Type A and Type B agents. Type A agents have stronger anchoring effect than the Type B agents. The simulation results reveal that the Type-A agents’ order price tends to uctuate more violently than the Type B agents, and that the Type A agents tend to lose more asset than the Type B.

Evaluation of automated-trading strategies using an artificial market

The purpose of this study was to extend the simulation of artificial markets to the practical application. We constructed an artificial-market system with an interface with the automated-trading strategies. Then, using an artificial-market simulation, we conducted two types of evaluations of automated-trading strategies that could participate in a Kaburobo competition. (1) Our system evaluated the risks and returns of the strategies in various market environments. An evaluation using the artificial market was able to provide better information than a conventional evaluation using a back test. (2) Our system could also test the market impact of automated-trading strategies. Our results revealed that the market impact of the strategies may not only depend on their rule content but also on the way they are combined with other strategies.

Artificial Market Simulation with Embedded Complex Network Structures

Artificial Market Simulation with Embedded Complex Network Structures