continuous double auction

We develop a behavioral model for liquidity and volatility based on empirical regularities in trading order flow in the London Stock Exchange. This can be viewed as a very simple agent based model in which all components of the model are validated against real data. Our empirical studies of order flow uncover several interesting regularities in the way trading orders are placed and cancelled. The resulting simple model of order flow is used to simulate price formation under a continuous double auction, and the statistical properties of the resulting simulated sequence of prices are compared to those of real data. The model is constructed using one stock (AZN) and tested on 24 other stocks. For low volatility, small tick size stocks (called Group I) the predictions are very good, but for stocks outside Group I they are not good. For Group I, the model predicts the correct magnitude and functional form of the distribution of the volatility and the bid-ask spread, without adjusting any parameters based on prices. This suggests that at least for Group I stocks, the volatility and heavy tails of prices are related to market microstructure effects, and supports the hypothesis that, at least on short time scales, the large fluctuations of absolute returns |r| are well described by a power law of the form P (|r| > R) ∼ R -αr , with a value of α r that varies from stock to stock. JEL: G10, General Financial Markets, volatility, bid-ask spread, behavioral finance A. Toward a more quantitative behavioral economics 1 We would like to thank the James S.

This paper studies the properties of the continuous double-auction trading mechanism using an artificial market populated by heterogeneous computational agents. In particular, we investigate how changes in the population of traders and in market microstructure characteristics affect price dynamics, information dissemination, and distribution of wealth across agents. In our computer-simulated market only a small fraction of the population observe the risky asset's fundamental value with noise, while the rest of the agents try to forecast the asset's price from past transaction data. In contrast to other artificial markets, we assume that the risky asset pays no dividend, thus agents cannot learn from past transaction prices and subsequent dividend payments. We find that private information can effectively disseminate in the market unless market regulation prevents informed investors from short selling or borrowing the asset, and these investors do not constitute a critical mass. In such case, not only are markets less efficient informationally, but may even experience crashes and bubbles. Finally, increased informational efficiency has a negative impact on informed agents' trading profits and a positive impact on artificial intelligent agents' profits.

Agent-Based Modelling is gaining wider acceptance as a paradigm for social research. However, it still present limitations in the management of the process to generate the simulations from the initial conceptual models. Thus, it is difficult to reuse the knowledge from available models and adapt it to different hypotheses. This paper proposes the use of metamodels in order to define explicitly the core concepts of a problem and to differentiate the aspects involved in the process. A case study for continuous double auctions shows how to define the related metamodel and use it to address alternative situations in these auctions. The case study drives the discussion on the advantages and limitations that metamodelling can bring to social simulation.

The "ZIP" adaptive trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets. The original ZIP algorithm requires the values of eight control parameters to be set correctly. A new extension of the ZIP algorithm, called ZIP60, requires the values of 60 parameters to be set correctly. ZIP60 is shown here to produce significantly better results than the original ZIP (called "ZIP8" hereafter), for negligable additional computational costs. A genetic algorithm (GA) is used to search the 60-dimensional ZIP60 parameter space, and it finds parameter vectors that yield ZIP60 traders with mean scores significantly better than those of ZIP8s. This paper shows that the optimizing evolutionary search works best when the GA itself controls the dimensionality of the search-space, so that the search commences in an 8-d space and thereafter the dimensionality of the search-space is gradually increased by the GA until it is exploring a 60-d space. Furthermore, the results from ZIP60 cast some doubt on prior ZIP8 results concerning the evolution of new 'hybrid' auction mechanisms that appeared to be better than the CDA.

The "ZIP" adaptive automated trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets populated by mixtures of human and "software robot" traders. Previous papers have shown that values of the eight parameters governing behavior of ZIP traders can be automatically optimized using a genetic algorithm (GA), and that markets populated by GA-optimized traders perform better than those populated by ZIP traders with manually-set parameter values. This paper introduces a more sophisticated version of the ZIP algorithm, called "ZIP60", which requires the values of 60 parameters to be set correctly. ZIP60 is shown here to produce significantly better results in comparison to the original ZIP algorithm (called "ZIP8" hereafter) when a GA is used to search the 60-dimensional parameter space. It is also demonstrated here that this works best when the GA itself has control over the dimensionality of the search-space, allowing evolution to guide the expansion of the search-space up from 8 parameters to 60 via intermediate steps. Principal component analysis of the best evolved ZIP60 parameter-sets establishes that no ZIP8 solutions are embedded in the 60-dimensional space. Moreover, some of the results and analysis presented here cast doubt on previously-published ZIP8 results concerning the evolution of new 'hybrid' auction mechanisms that appeared to be improvements on the CDA: it now seems likely that those results were actually consequences of the relative lack of sophistication in the original ZIP8 algorithm, because "hybrid" mechanisms occur much less frequently when ZIP60s are used.

This paper describes the use of a genetic algorithm (GA) to find optimal parametervalues for trading agents that operate in virtual online auction "e-marketplaces", where the rules of those marketplaces are also under simultaneous control of the GA. The aim is to use the GA to automatically design new mechanisms for agent-based e-marketplaces that are more efficient than online markets designed by (or populated by) humans. The space of possible auction-types explored by the GA includes the Continuous Double Auction (CDA) mechanism (as used in most of the world's financial exchanges), and also two purely one-sided mechanisms. Surprisingly, the GA did not always settle on the CDA as an optimum. Instead, novel hybrid auction mechanisms were evolved, which are unlike any existing market mechanisms. In this paper we show that, when the market supply and demand schedules undergo sudden "shock" changes partway through the evaluation process, two-sided hybrid market mechanisms can evolve which may be unlike any human-designed auction and yet may also be significantly more efficient than any human designed market mechanism.

algorithmic mechanism design, auction and negotiation technology, automated trading, ZIP traders, genetic algorithms, e-marketplaces This paper builds on previous papers describing our ongoing research in automated market-mechanism design: using a genetic algorithm (GA) to find optimal parametersettings for software-agent traders that operate in virtual "e-marketplaces", where the rules of the marketplaces are also under simultaneous control of the GA. The aim is that the GA automatically designs new agent-based e-marketplaces that are more efficient than existing markets designed by (or populated by) humans. Das et al. recently demonstrated that ZIP software-agent traders consistently out-perform human traders in Continuous Double Auction (CDA) marketplaces similar to those used in the international financial markets. Cliff used a GA to explore a continuous space of ZIP-trader auction-market mechanisms, where the space of possible auction-types explored included the CDA and also two purely one-sided mechanisms. Surprisingly, the GA would sometimes settle on novel hybrid auction mechanisms partway between the CDA and a one-sided auction. Such results occurred when the market's supply and demand schedules were unchanging, and also when the schedules undergo a single sudden "shock" change halfway through the evaluation process. These results could prima facie support the hypothesis that hybrid auctions are in general preferable to the well-known CDA mechanism. In this paper we present new results that clarify that hypothesis. In our new experiments, more than one shock-change in supply/demand occurs during the mechanism-evaluation process; under this regimen, CDA auction-mechanisms are identified by the GA as optimal in three of the four experiments reported here, and in the fourth experiment a near-CDA hybrid mechanism was evolved. From these results we conclude that, while evolved hybrid market-mechanisms may be useful in niches where the marketplace's supply and demand dynamics are known a priori to be relatively stable or regular, the CDA remains the mechanism of choice when it is difficult or impossible to make accurate predictions concerning the dynamic stability of the supply and demand schedules.

algorithmic trading, online auction marketplaces, emarketplaces, automated market mechanism design, trader-agents, ZIP traders, genetic algorithms The Zero-Intelligence Plus ("ZIP") adaptive automated trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets populated by mixtures of human and "robot" traders . To successfully populate a market with ZIP traders, the values of eight control parameters need to be set correctly. While these eight values can be set manually, previous papers have demonstrated that values of those parameters can be automatically optimized using a genetic algorithm (GA), to tailor ZIP traders to particular markets, and also (by adding an additional real-valued numeric parameter) to automatically discover novel new forms of auction market mechanism that are more efficient than the CDA. This paper introduces a more sophisticated version of the ZIP algorithm, which is shown to produce significantly better results. The extended variant is known as "ZIP60", because it requires 60 real-valued control parameters to be set correctly, and the original ZIP algorithm is re-named "ZIP8" accordingly. Manually choosing the correct values for 60 control parameters would be a very laborious task, but it is demonstrated here that an appropriate automatic optimization process can discover good sets of values for the parameters. A simple GA operating in the 60dimensional parameter space is shown to produce ZIP60 traders with mean scores significantly improved over ZIP8s, but also with high variance in those improvements. A slight revision of this approach is shown to give results with even higher mean improvements and also with lower variance in those improvements. The revised approach involves giving the GA control over the dimensionality of the parameter space being searched, starting with an eight-dimensional space and then allowing the GA to automatically and gradually expand that space up to sixtydimensional only when the increased number of parameters leads to identifiably better solutions. The results from ZIP60, while better than ZIP8, show a greatly reduced incidence of cases where the GA discovers auction mechanisms that are significantly better than the fixed CDA mechanism. This may be interpretable as evidence that the earlier ZIP8 results (where improvements on the CDA were common) were consequences of the relative lack of sophistication in the ZIP8 algorithm.

algorithmic mechanism design, auction and negotiation technology, automated trading, ZIP traders, genetic algorithms, e-marketplaces This paper builds on previous papers describing our ongoing research in automated market-mechanism design: using a genetic algorithm (GA) to find optimal parametersettings for software-agent traders that operate in virtual "e-marketplaces", where the rules of the marketplaces are also under simultaneous control of the GA. The aim is that the GA automatically designs new agent-based e-marketplaces that are more efficient than existing markets designed by (or populated by) humans. Das et al. recently demonstrated that ZIP software-agent traders consistently out-perform human traders in Continuous Double Auction (CDA) marketplaces similar to those used in the international financial markets. Cliff used a GA to explore a continuous space of ZIP-trader auction-market mechanisms, where the space of possible auction-types explored included the CDA and also two purely one-sided mechanisms. Surprisingly, the GA would sometimes settle on novel hybrid auction mechanisms partway between the CDA and a one-sided auction. Such results occurred when the market's supply and demand schedules were unchanging, and also when the schedules undergo a single sudden "shock" change halfway through the evaluation process. These results could prima facie support the hypothesis that hybrid auctions are in general preferable to the well-known CDA mechanism. In this paper we present new results that clarify that hypothesis. In our new experiments, more than one shock-change in supply/demand occurs during the mechanism-evaluation process; under this regimen, CDA auction-mechanisms are identified by the GA as optimal in three of the four experiments reported here, and in the fourth experiment a near-CDA hybrid mechanism was evolved. From these results we conclude that, while evolved hybrid market-mechanisms may be useful in niches where the marketplace's supply and demand dynamics are known a priori to be relatively stable or regular, the CDA remains the mechanism of choice when it is difficult or impossible to make accurate predictions concerning the dynamic stability of the supply and demand schedules.

Gode and Sunder's (1993) results from using "zero-intelligence" (zi) traders, that act randomly within a structured market, appear to imply that convergence to the theoretical equilibrium price in continuous double-auction markets is determined more by market structure than by the intelligence of the traders in that market. However, it is demonstrated here that the average transaction prices of zi traders can vary significantly from the theoretical equilibrium value when the market supply and demand are asymmetric, and that the degree of difference from equilibrium is predictable from a priori probabilistic analysis. In this sense, it is shown here that Gode and Sunder's results are artefacts of their experimental regime. Following this, `zero-intelligence-plus' (zip) traders are introduced: like zi traders, these simple agents make stochastic bids. Unlike zi traders, they employ an elementary form of machine learning. Groups of zip traders interacting in exper...

We develop a new model to analyse the strategic behaviour of buyers and sellers in market mechanisms. In particular, we wish to understand how the different strategies they adopt affect their economic efficiency in the market and to understand the impact of these choices on the overall efficiency of the marketplace. To this end, we adopt a two-population evolutionary game theoretic approach, where we consider how the behaviours of both buyers and sellers evolve in marketplaces. In so doing, we address the shortcomings of the previous state-of-the-art analytical model that assumes that buyers and sellers have to adopt the same mixed strategy in the market. Finally, we apply our model in one of the most common market mechanisms, the Continuous Double Auction, and demonstrate how it allows us to provide new insights into the strategic interactions of such trading agents.

The "ZIP" adaptive automated trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets populated by mixtures of human and "software robot" traders. Previous papers have shown that values of the eight parameters governing behavior of ZIP traders can be automatically optimized using a genetic algorithm (GA), and that markets populated by GA-optimized traders perform better than those populated by ZIP traders with manually-set parameter values. This paper introduces a more sophisticated version of the ZIP algorithm, called "ZIP60", which requires the values of 60 parameters to be set correctly. ZIP60 is shown here to produce significantly better results in comparison to the original ZIP algorithm (called "ZIP8" hereafter) when a GA is used to search the 60-dimensional parameter space. It is also demonstrated here that this works best when the GA itself has control over the dimensionality of the search-space, allowing evolution to guide the expansion of the search-space up from 8 parameters to 60 via intermediate steps. Principal component analysis of the best evolved ZIP60 parameter-sets establishes that no ZIP8 solutions are embedded in the 60-dimensional space. Moreover, some of the results and analysis presented here cast doubt on previously-published ZIP8 results concerning the evolution of new 'hybrid' auction mechanisms that appeared to be improvements on the CDA: it now seems likely that those results were actually consequences of the relative lack of sophistication in the original ZIP8 algorithm, because "hybrid" mechanisms occur much less frequently when ZIP60s are used.

This paper describes the use of a genetic algorithm (GA) to find optimal parameter-values for trading agents that operate in virtual online auction "e-marketplaces", where the rules of those marketplaces are also under simultaneous control of the GA. The aim is to use the GA to automatically design new mechanisms for agent-based e-marketplaces that are more efficient than online markets designed by (or populated by) humans. The space of possible auction-types explored by the GA includes the Continuous Double Auction (CDA) mechanism (as used in most of the world's financial exchanges), and also two purely one-sided mechanisms. Surprisingly, the GA did not always settle on the CDA as an optimum. Instead, novel hybrid auction mechanisms were evolved, which are unlike any existing market mechanisms. In this paper we show that, when the market supply and demand schedules undergo sudden "shock" changes partway through the evaluation process, two-sided hybrid market mechanisms can evolve which may be unlike any human-designed auction and yet may also be significantly more efficient than any human designed market mechanism.

The continuous double-auction (CDA) is a powerful market mechanism, noted for its speed and efficiency, and is the mechanism underlying the organization of open-outcry 'trading pits' at major international derivatives markets. In previous publications, Cliff & Bruten demonstrated that software 'trading agents' need more than zero intelligence to give human-like CDA price-equilibration behavior and presented results from experiments with simple adaptive trading agents in CDA markets and in one-sided auctions. These agents give very good performance on standard measures of trading activity such as allocative efficiency, Smith's 'alpha' measure of price convergence, and profit dispersion, but only when parameters governing the adaptation mechanism are set to appropriate values. Determining good or optimal combinations of parameters by hand is possible, but can be labor-intensive. This paper presents the first results from using a genetic algorithm to optimize all the real-valued parameters governing adaptation in the trading agents. It is shown that a simple genetic algorithm (GA), in combination with an appropriate evaluation function, can deliver good parameter settings from random initial-value conditions. The evolutionary trajectories of the population through the 8 -dimensional parameter space are illustrated, and the use of the GA to identify parameters that are redundant or even harmful is discussed.

Simulation experiments are conducted on simple continuous double auction (CDA) markets based on the experimental economics work of Vernon Smith. CDA models within experimental economics usually consist of a sequence of discrete trading periods or "days", with allocations of stock and currency replenished at the start of each day, a situation we call "periodic" replenishment. In our experiments we look at both periodic and continuous-replenishment versions of the CDA. In this we build on the work of Cliff and Preist ( ) with human subjects, but we replace human traders with Zero Intelligence Plus (ZIP) trading agents, a minimal algorithm that can produce equilibrating market behaviour in CDA models. Our results indicate that continuous-replenishment (CR) CDA markets are similar to conventional periodic CDA markets in their ability to show equilibration dynamics. Secondly we show that although both models produce the same behaviour of price formation, they are different playing fields, as periodic markets are more efficient over time than their continuous counterparts. We also find, however, that the volume of trade in periodic CDA markets is concentrated in the early period of each trading day, and the market is in this sense inefficient. We look at whether ZIP agents require different parameters for optimal behaviour in each market type, and find that this is indeed the case. Overall, our conclusions mirror earlier findings on the robustness of the CDA, but we stress that a CR-CDA marketplace equilibrates in a different way to a periodic one.

Designing market-based mechanism that benefits both the cloud customer and cloud provider in a cloud market is a fundamental but complex problem. Double auction is one such mechanism to allocate resources that prevents monopoly and is used to design an unbiased optimal market strategy for cloud market. This work proposes a truthful combinatorial double auction mechanism for allocation and pricing of computing resources in cloud. For resource allocation, utilitarian social welfare maximization problem is formulated using Integer Linear Programming (ILP) and a near optimal solution is obtained using Linear Programming based padded method. For payment, truthful and novel schemes are designed for both customers and providers. Moreover, the proposed mechanism is individual rational, computationally tractable, weakly budget-balance and asymptotic efficient. Performance evaluation and comparative study exhibit that the proposed mechanism is effective on various performance metrics such as utilitarian social welfare, total utility, customers' satisfaction, providers' revenue and hence is applicable in real cloud environments.

Töös rakendatakse ühte tellimusraamatu stohhastilist mudelit AstraZeneca PLC aktsiahinna muutumise kirjeldamiseks päevasisese kauplemise tingimustes. Vaadeldava mudeli korral on võimalik saada mitmetele huvipakkuvatele tellimusraamatu käitumisega setud küsimustele analüütilisel kujul vastused, mis on oluliselt efektiivsem vastuste leidmisest Monte-Carlo simulatsioonide abil. Laplace teisenduse meetodi rakendamise teel saadakse valemid mitmete huvipakkuvate sündmuste tinglike jaotuste jaoks. Töös käsitletakse põhjalikumalt turuhinna kasvamise tõenäosuse arvutamist ning parima ostuhinnaga tehingutellimuse ennem turuhinna muutumist täitmise tõenäosuse leidmist. Võrdlus sündmuste empiiriliste toimumissagedustega näitab, et vaadeldav mudel kirjeldab küllalt hästi vaadeldava aktsia tellimusraamatu käitumise dünaamikat lühikese ajahorisondi korral.

Töös rakendatakse ühte tellimusraamatu stohhastilist mudelit AstraZeneca PLC aktsiahinna muutumise kirjeldamiseks päevasisese kauplemise tingimustes. Vaadeldava mudeli korral on võimalik saada mitmetele huvipakkuvatele tellimusraamatu käitumisega setud küsimustele analüütilisel kujul vastused, mis on oluliselt efektiivsem vastuste leidmisest Monte-Carlo simulatsioonide abil. Laplace teisenduse meetodi rakendamise teel saadakse valemid mitmete huvipakkuvate sündmuste tinglike jaotuste jaoks. Töös käsitletakse põhjalikumalt turuhinna kasvamise tõenäosuse arvutamist ning parima ostuhinnaga tehingutellimuse ennem turuhinna muutumist täitmise tõenäosuse leidmist. Võrdlus sündmuste empiiriliste toimumissagedustega näitab, et vaadeldav mudel kirjeldab küllalt hästi vaadeldava aktsia tellimusraamatu käitumise dünaamikat lühikese ajahorisondi korral.

To understand the impact of high frequency trading (HFT) systems on financial market dynamics, a series of controlled real-time experiments involving humans and automated trading agents were performed. These experiments fall at the interdisciplinary boundary between the more traditional fields of behavioural economics (human-only experiments) and agent based computational economics (agent-only simulations). Experimental results demonstrate that: (a) faster financial trading agents can reduce market efficiency-a worrying result given the race towards zerolatency (ever faster trading) observed in real markets; and (b) faster agents can lead to market fragmentation, such that markets transition from a regime where humans and agents freely interact, to a regime where agents are more likely to trade between themselves-a result that has also been observed in real financial markets. It is also shown that (c) realism in experimental design can significantly alter market dynamics-suggesting that, if we want to understand complexity in real financial markets, it is finally time to move away from the simple experimental economics models first introduced in the 1960s.

In this paper, we describe a novel bidding strategy that autonomous trading agents can use to participate in Continuous Double Auctions (CDAs). Our strategy is based on both short and long-term learning that allows such agents to adapt their bidding behaviour to be efficient in a wide variety of environments. For the shortterm learning, the agent updates the aggressiveness of its bidding behaviour (more aggressive means it will trade off profit to improve its chance of transacting, less aggressive that it targets more profitable transactions and is willing to trade off its chance of transacting to achieve them) based on market information observed after any bid or ask appears in the market. The long-term learning then determines how this aggressiveness factor influences an agent's choice of which bids or asks to submit in the market, and is based on market information observed after every transaction (successfully matched bid and ask). The principal motivation for the short-term learning is to enable the agent to immediately respond to market fluctuations, while for the long-term learning it is to adapt to broader trends in the way in which the market demand and supply changes over time. We benchmark our strategy against the current state of the art (ZIP and GDX) and show that it outperforms these benchmarks in both static and dynamic environments. This is true both when the population is homogeneous (where the increase in efficiency is up to 5.2%) and heterogeneous (in which case there is a 0.85 probability of our strategy being adopted in a two-population evolutionary game theoretic analysis).

CDA, automated mechanism design, ZI-C Agents, hybrid markets, agent-based simulation, evolutionary strategy, BICAS The Continuous Double Auction (CDA) is one of the most popular of all auction markets in the world. Some of the biggest trade markets in the world including the New York Stock Exchange (NYSE) and the Chicago Mercantile Exchange are organised as CDAs. Previous work by Cliff has shown that previously unknown 'hybrid' variants of the traditional CDA can be found to give the most desirable market dynamics. Cliff's results were based on experiments conducted using a computational simulation of the CDA populated by electronic Zero Intelligence Plus (ZIP) traders and his work uses a GA to co-evolve the market mechanism as well as the ZIP agent parameters. The exclusive use of the ZIP trading algorithm in all his work raises questions about the robustness of his results to any change in the trading algorithm used. In this thesis we use a self-adaptive Evolutionary Strategy (ES) to explore the space of possible auction types in a CDA populated by Gode and Sunder's cognitively simple Zero Intelligence (ZI) traders. We show that hybrid CDAs are still preferred over traditional variants and our results provide the first demonstration that hybrid variants of the CDA can provide favourable dynamics for trading strategies other than ZIP.

In this paper, we describe a novel bidding strategy that autonomous trading agents can use to participate in Continuous Double Auctions (CDAs). Our strategy is based on both short and long-term learning that allows such agents to adapt their bidding behaviour to be efficient in a wide variety of environments. For the shortterm learning, the agent updates the aggressiveness of its bidding behaviour (more aggressive means it will trade off profit to improve its chance of transacting, less aggressive that it targets more profitable transactions and is willing to trade off its chance of transacting to achieve them) based on market information observed after any bid or ask appears in the market. The long-term learning then determines how this aggressiveness factor influences an agent's choice of which bids or asks to submit in the market, and is based on market information observed after every transaction (successfully matched bid and ask). The principal motivation for the short-term learning is to enable the agent to immediately respond to market fluctuations, while for the long-term learning it is to adapt to broader trends in the way in which the market demand and supply changes over time. We benchmark our strategy against the current state of the art (ZIP and GDX) and show that it outperforms these benchmarks in both static and dynamic environments. This is true both when the population is homogeneous (where the increase in efficiency is up to 5.2%) and heterogeneous (in which case there is a 0.85 probability of our strategy being adopted in a two-population evolutionary game theoretic analysis).

Economic models are found efficient in managing heterogeneous computer resources such as storage, CPU and memory for grid computing. Commodity market, double auction and contract-net-protocol economic models have been widely discussed in the literature. These models are suitable for sharing distributed computer resources that belong to different owners. Agent technology can be used to manage these heterogeneous resources without human intervention, since agents are autonomous and intelligent in behavior. In this paper, we develop and simulate an agent-oriented double auction economic model. We compare the performance of our agent-oriented model with traditional double auction model, and show that the agent-oriented model is good in maximizing profit for providers.

We develop a novel bidding strategy that software agents can use to buy and sell goods in Continuous Double Auctions (CDAs). Our strategy involves the agent forming a bid or ask by assessing the degree of risk involved and making a prediction about the competitive equilibrium that is likely to be reached in the marketplace. We benchmark our strategy against two of the most common strategies for CDAs, namely the Zero-Intelligence and the Zero-Intelligence Plus strategies, and we show that our agents outperform these benchmarks. Specifically, our agents win in 100% of the simulations against the ZI agents and, on average, 75% of the games against the ZIP agents.

This paper reports on the design and comparison of two economically inspired mechanisms for task allocation in environments where sellers have finite production capacities and a cost structure composed of a fixed overhead cost and a constant marginal cost. Such mechanisms are required when a system consists of multiple self-interested stakeholders that each possess private information that is relevant to solving a systemwide problem. Against this background, we first develop a computationally tractable centralized mechanism that finds the set of producers that have the lowest total cost in providing a certain demand (i.e., it is efficient). We achieve this by extending the standard Vickrey-Clarke-Groves mechanism to allow for multiattribute bids and by introducing a novel penalty scheme such that producers are incentivized to truthfully report their capacities and their costs. Furthermore, our extended mechanism is able to handle sellers' uncertainty about their production capacity and ensures that individual agents find it profitable to participate in the mechanism. However, since this first mechanism is centralized, we also develop a complementary decentralized mechanism based around the continuous double auction. Again, because of the characteristics of our domain, we need to extend the standard form of this protocol by introducing a novel clearing rule based around an order book. With this modified protocol, we empirically demonstrate (with simple trading strategies) that the mechanism achieves high efficiency. In particular, despite this simplicity, the traders can still derive a profit from the market which makes our mechanism attractive since these results are a likely lower bound on their expected returns. Index Terms-Decision theory, distributed decision making, market-based control (MBC), multiagent systems. I. INTRODUCTION T ASK allocation is an important and challenging problem for computer science (see [5] and [36] for an overview). To this end, in this paper, we specifically consider it in the context of assigning tasks to a set of autonomous software agents. Now, when a designer has complete control over both the agents and the way in which they interact, an inherently cooperative approach can be developed whereby the agents work together Manuscript

We develop a novel bidding strategy that software agents can use to buy and sell goods in Continuous Double Auctions (CDAs). Our strategy involves the agent forming a bid or ask by assessing the degree of risk involved and making a prediction about the competitive equilibrium that is likely to be reached in the marketplace. We benchmark our strategy against two of the most common strategies for CDAs, namely the Zero-Intelligence and the Zero-Intelligence Plus strategies, and we show that our agents outperform these benchmarks. Specifically, our agents win in 100% of the simulations against the ZI agents and, on average, 75% of the games against the ZIP agents.

Creating a Wordle game in Google Sheets, with ChatGPT was a real back-and-forth. ChatGPT, with its vast pool of knowledge and ideas, was a huge help. But in the end, it was the practical skills-knowing what makes a good user experience and applying design thinking-that shaped the project. Every iteration brought new challenges and things to learn, highlighting the value of patience, persistence, and paying attention to the small stuff in making digital tools that people might enjoy using.

This paper reports a laboratory experiment that studies several features of a tradable emission permit program recently implemented in the Los Angeles area. The experiment focuses on the new Electronic Bulletin Board trading institution, in which firms publicly post proposed terms of trade. Potential trading partners can review this information online, and transactions are executed following bilateral negotiation. The experiment includes trading restrictions implemented in the regulations due to the geography of Los Angeles. We find that the bulletin board market performs well and that prices reflect market conditions as accurately as in the continuous double auction trading institution. 1 Financial support was provided by the Southern California Studies Center. We are grateful to seminar participants at the USC School of Urban and Regional Planning, Resources for the Future and Emory University, participants at the Public Choice Society/Economic Science Association conference, Jeffrey Nugent and an anonymous referee for valuable comments. 56 TIMOTHY N. CASON AND LATA GANGADHARAN 2 One drawback of tradable permit programs in general is that without geographic restrictions, tradable permits can result in concentrated emission "hot spots." This is of little consequence for global environmental problems such as carbon emissions; however, for other environmental problems such as this the emission location is important.

We study consequences of regulatory interventions in limit order markets that aim at stabilizing the market after an occurrence of a "flash crash". We use a simulation platform that creates random arrivals of trade orders, that allows us to analyze subtle features of liquidity and price variability under various market structures. The simulations are performed under continuous double-auction microstructure, and under alternatives, including imposing minimum resting times, shutting off trading for a period of time, and switching to call auction mechanisms. We find that the latter is the most effective in restoring the liquidity of the book and recovery of the price level. However, one has to be cautious about possible long term consequences of the intervention on the traders' strategies.

Agent-Based Modelling is gaining wider acceptance as a paradigm for social research. However, it still present limitations in the management of the process to generate the simulations from the initial conceptual models. Thus, it is difficult to reuse the knowledge from available models and adapt it to different hypotheses. This paper proposes the use of metamodels in order to define explicitly the core concepts of a problem and to differentiate the aspects involved in the process. A case study for continuous double auctions shows how to define the related metamodel and use it to address alternative situations in these auctions. The case study drives the discussion on the advantages and limitations that metamodelling can bring to social simulation.

Prediction markets are a promising instrument for drawing on the "wisdom of the crowds". For instance, in a corporate context they have been used successfully to forecast sales or project risks by tapping into the heterogeneous information of decentralized actors in and outside of companies. Among the main market mechanisms implemented so far in prediction markets are () the continuous double auction and () the logarithmic market scoring rule. However, it is not fully understood how this choice a ects crucial variables like prediction market accuracy or price variation. Our paper uses an experiment-based and micro validated simulation model to improve the understanding of the mechanism-related e ects and to inform further laboratory experiments. The results underline the impact of mechanism selection. Due to the higher number of trades and the lower standard deviation of the price, the logarithmic market scoring rule seems to have a clear advantage at a first glance. This changes when the accuracy level, which is the most important criterion from a practical perspective, is used as an independent variable; the e ects become less straightforward and depend on the environment and actors. Besides these contributions, this work provides an example of how experimental data can be used to validate agent strategies on the micro level using statistical methods.

In this paper, we introduce a continuous double auction method for grid resource allocation in which resources are considered as provider agents and users as consumer agents. In each time step, each provider agent determines its requested value based on its workload and each consumer agent determines its bid value based on two constraints: the remaining time for bidding, and the remaining resources for bidding. We study this method in terms of economic efficiency and system performance. Experimental results show that the proposed method is better than Earliest Deadline First (EDF) method, which is a default strategy in many schedulers.

A computational grid is composed of a set of resource consumers and resources providers. Usually these entities are independent and making decisions autonomously based on their policies and resource allocation in such systems is a challenging problem. In such systems using market-like techniques for this problem regulates the supply and demand for resources, provides an incentive for providers, and motivates the users to trade-off between deadline, budget, and the required level of quality of service. In this paper, we introduce a continuous double auction method (CDA) for grid resource allocation in which resources are considered as provider agents and users as consumer agents. In our proposed method these entities are allowed to participate in a grid independently and make decisions autonomously. We study this method in terms of economic efficiency and system performance. Experimental results illustrate that the proposed method is efficient in terms of successful execution rates, resource utilization rates and fair profit allocation.

and Tivnan (2012) recently suggested the existence of a phase transition in the dynamics of financial markets in which there is free interaction between human traders and algorithmic trading systems ('robots'). Above a particular time-threshold, humans and robots trade with one another; below the threshold all transactions are robot-to-robot. We refer to this abrupt system transition as the 'robot phase transition'. Here, we conduct controlled experiments where human traders interact with 'robot' trading agents in minimal models of electronic financial markets to see if correlates of the two regimes suggested by Johnson et al. (2012) occur in such laboratory conditions. Our results indicate that when trading robots act on a superhuman timescale, the market starts to fragment, with statistically lower human-robot interactions than we would expect from a fully mixed market. We tentatively conclude that this is the first empirical evidence for the robot phase transition occurring under controlled experimental conditions.

We report on results from experiments where human traders interact with software-agent traders in a real-time asynchronous continuous double auction (CDA) experimental economics system. Our experiments are inspired by the seminal work reported by IBM at IJCAI 2001 [Das et al., 2001], where it was demonstrated that software-agent traders could consistently outperform human traders in real-time CDA markets. IBM tested two trading-agent strategies, ZIP and a modified version of GD, and in a subsequent paper they reported on a new strategy called GDX that was demonstrated to outperform GD and ZIP in agent vs. agent CDA competitions, on which basis it was claimed that GDX "... may offer the best performance of any published CDA bidding strategy." [Tesauro and Bredin, 2002]. In this paper, we employ experiment methods similar to those pioneered by IBM to test the performance of "Adaptive Aggressive" (AA) algorithmic traders [Vytelingum, 2006]. The results presented her...

In 2001, a team of researchers at IBM published a paper in IJCAI which reported on the first experiments that systematically studied the interactions of human traders and software-agent traders in electronic marketplaces running the continuous double auction (CDA) mechanism. IBM found that two software-agent strategies, known as GD and ZIP, consistently outperformed human traders. IBM's results received international press coverage, probably because the CDA is the mechanism that is used in the main electronic trading systems that make up the global financial markets. In 2002, Tesauro & Bredin published details of an extension to GD, which they named GDX, for which they wrote: "We suggest that this algorithm may offer the best performance of any published CDA bidding strategy". To the best of our knowledge, GDX has never been tested against human traders under experimental conditions. In this paper, we report on the first such test: we present detailed analysis of the results from our own replications of IBM's human vs. ZIP experiments and from our world-first experiments that test humans vs. GDX. Our overall findings are that, both when competing against ZIP in pure agent vs. agent experiments and when competing against human traders, GDX's performance is significantly better than the performance of ZIP.

We present results demonstrating that an appropriately configured deep learning neural network (DLNN) can automatically learn to be a high-performing algorithmic trading system, operating purely from training-data inputs generated by passive observation of an existing successful trader T. That is, we can point our black-box DLNN system at trader T and successfully have it learn from T's trading activity, such that it trades at least as well as T. Our system, called DeepTrader, takes inputs derived from Level-2 market data, i.e. the market's Limit Order Book (LOB) or Ladder for a tradeable asset. Unusually, DeepTrader makes no explicit prediction of future prices. Instead, we train it purely on input-output pairs where in each pair the input is a snapshot S of Level-2 LOB data taken at the time when T issued a quote Q (i.e. a bid or an ask order) to the market; and DeepTrader's desired output is to produce Q when it is shown S. That is, we train our DLNN by showing it the LOB data S that T saw at the time when T issued quote Q, and in doing so our system comes to behave like T, acting as an algorithmic trader issuing specific quotes in response to specific LOB conditions. We train DeepTrader on large numbers of these S/Q snapshot/quote pairs, and then test it in a variety of market scenarios, evaluating it against other algorithmic trading systems in the public-domain literature, including two that have repeatedly been shown to outperform human traders. Our results demonstrate that DeepTrader learns to match or outperform such existing algorithmic trading systems. We analyse the successful DeepTrader network to identify what features it is relying on, and which features can be ignored. We propose that our methods can in principle create an explainable copy of an arbitrary trader T via "black-box" deep learning methods.

A 24-hour exchange market was created on the Web to trade political futures contracts using fictitious money. In this online market, a political futures contract is a futures contract which matures on the election day with a liquidation price determined by the percentage of votes a candidate receives on the election day. Continuous double auctions were implemented as the system for order storage and price discovery. We drew market participants in the form of tournaments in which top traders won cash awards. Such a market was run, with about 400 registered traders, during the U.S. presidential election in November 2004 and Taiwan parliamentary election in December 2004. The experiments recorded transaction price, highest bid, lowest ask, and trading volume of each contract as a function of time. Despite the relatively small scale of the exchange, in terms of the number of participants and duration of the tournament, we report evidence for asymptotic power-law behaviors of the distributions of price returns, trading volumes, inter-transaction time intervals, and accumulated wealth that were found universal in real financial markets.

We propose a novel variant of the Continuous Double Auction (CDA), the Trust-based CDA (T-CDA), which we demonstrate to be robust to execution uncertainty. This is desirable in a setting where traders may fail to deliver the goods, services or payments they have promised. Specifically, the T-CDA provides a mechanism that allows agents to commit to trades they believe will maximize their expected utility. In this paper, we consider agents that use their trust in other agents to estimate the expected utility of a transaction. We empirically evaluate the mechanism, both against the optimal solution given perfect and complete information and against the standard CDA. We show that the T-CDA consistently outperforms the traditional CDA as execution uncertainty increases in the system. Furthermore, we investigate the robustness of the mechanism to unreliable trust information and find that performance degrades gracefully as information quality decreases. * This paper presents work done while the author was at the University of Southampton.

Agent-Based Modelling is gaining wider acceptance as a paradigm for social research. However, it still present limitations in the management of the process to generate the simulations from the initial conceptual models. Thus, it is difficult to reuse the knowledge from available models and adapt it to different hypotheses. This paper proposes the use of metamodels in order to define explicitly the core concepts of a problem and to differentiate the aspects involved in the process. A case study for continuous double auctions shows how to define the related metamodel and use it to address alternative situations in these auctions. The case study drives the discussion on the advantages and limitations that metamodelling can bring to social simulation.

Agent-Based Modelling is gaining wider acceptance as a paradigm for social research. However, it still present limitations in the management of the process to generate the simulations from the initial conceptual models. Thus, it is difficult to reuse the knowledge from available models and adapt it to different hypotheses. This paper proposes the use of metamodels in order to define explicitly the core concepts of a problem and to differentiate the aspects involved in the process. A case study for continuous double auctions shows how to define the related metamodel and use it to address alternative situations in these auctions. The case study drives the discussion on the advantages and limitations that metamodelling can bring to social simulation.

Agent-Based Modelling is gaining wider acceptance as a paradigm for social research. However, it still present limitations in the management of the process to generate the simulations from the initial conceptual models. Thus, it is difficult to reuse the knowledge from available models and adapt it to different hypotheses. This paper proposes the use of metamodels in order to define explicitly the core concepts of a problem and to differentiate the aspects involved in the process. A case study for continuous double auctions shows how to define the related metamodel and use it to address alternative situations in these auctions. The case study drives the discussion on the advantages and limitations that metamodelling can bring to social simulation.

Agent-Based Modelling is gaining wider acceptance as a paradigm for social research. However, it still present limitations in the management of the process to generate the simulations from the initial conceptual models. Thus, it is difficult to reuse the knowledge from available models and adapt it to different hypotheses. This paper proposes the use of metamodels in order to define explicitly the core concepts of a problem and to differentiate the aspects involved in the process. A case study for continuous double auctions shows how to define the related metamodel and use it to address alternative situations in these auctions. The case study drives the discussion on the advantages and limitations that metamodelling can bring to social simulation.

We develop a behavioral model for liquidity and volatility based on empirical regularities in trading order flow in the London Stock Exchange. This can be viewed as a very simple agent based model in which all components of the model are validated against real data. Our empirical studies of order flow uncover several interesting regularities in the way trading orders are placed and cancelled. The resulting simple model of order flow is used to simulate price formation under a continuous double auction, and the statistical properties of the resulting simulated sequence of prices are compared to those of real data. The model is constructed using one stock (AZN) and tested on 24 other stocks. For low volatility, small tick size stocks (called Group I) the predictions are very good, but for stocks outside Group I they are not good. For Group I, the model predicts the correct magnitude and functional form of the distribution of the volatility and the bid-ask spread, without adjusting any parameters based on prices. This suggests that at least for Group I stocks, the volatility and heavy tails of prices are related to market microstructure effects, and supports the hypothesis that, at least on short time scales, the large fluctuations of absolute returns |r| are well described by a power law of the form P (|r| > R) ∼ R −αr , with a value of α r that varies from stock to stock. JEL: G10, General Financial Markets, volatility, bid-ask spread, behavioral finance I. MOTIVATION AND BACKGROUND A. Toward a more quantitative behavioral economics 1 In the last two decades the field of behavioral finance has presented many examples where equilibrium rational choice models are not able to explain real economic behavior 2

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.

RESUMEN En este trabajo se estudia la formación de precios en un mercado artificial de doble subasta continua con agentes heterogéneos, tanto en términos de eficiencia informativa como en términos de sus propiedades estadísticas. A diferencia de otros mercados artificiales propuestos en la literatura, en este mercado existe información asimétrica tanto ex-ante como ex-post, puesto que los agentes no informados observan únicamente precios de transacción pasados. En consecuencia, su capacidad predictiva sobre el proceso fundamental está limitada por el grado de eficiencia informativa de los precios de transacción, endógena al mercado. Nuestro mercado es capaz de replicar los hechos estilizados observados comúnmente en las series reales de rendimientos (colas gruesas, persistencia en la volatilidad, correlación serial y efectos ARCH), y nos permite extraer predicciones teóricas con respecto al efecto de la distribución de la población de agentes sobre dichas propiedades estadísticas. Una diferencia del trabajo es la modelización del precio fundamental como un proceso estocástico, lo cual permite calibrar los parámetros del mercado artificial a datos reales. La principal conclusión del trabajo es que el mecanismo de doble subasta continua permite un elevado grado de eficiencia informativa cuando existe información asimétrica, además se observa que la eficiencia del mercado puede mejorar introduciendo inversores sin información privilegiada pero que explotan la información contenida en los precios de transacción pasados (por ejemplo, analistas técnicos y agentes con capacidad de aprendizaje), aunque ello disminuya el valor de la información privada y por tanto la riqueza de aquellos con información privilegiada.

The main goal of this paper is to show how relatively minor modifications of well-known algorithms (in particular, back propagation) can dramatically increase the performance of an artificial neural network (ANN) for time series prediction. We denote our proposed sets of modifications as the 'self-momentum', 'Freud' and 'Jung' rules. In our opinion, they provide an example of an alternative approach to the design of learning strategies for ANNs, one that focuses on basic mathematical conceptualization rather than on formalism and demonstration. The complexity of actual prediction problems makes it necessary to experiment with modelling possibilities whose inherent mathematical properties are often not well understood yet. The problem of time series prediction in stock markets is a case in point. It is well known that asset price dynamics in financial markets are difficult to trace, let alone to predict with an operationally interesting degree of accuracy. We therefore take financial prediction as a meaningful test bed for the validation of our techniques. We discuss in some detail both the theoretical underpinnings of the technique and our case study about financial prediction, finding encouraging evidence that supports the theoretical and operational viability of our new ANN specifications. Ours is clearly only a preliminary step. Further developments of ANN architectures with more and more sophisticated 'learning to learn' characteristics are now under study and test.

Autonomous agents are widely applied to automate interactions in robotics, e.g. for selling and purchasing goods on eBay, and in financial markets, e.g. in the form of quote machines and algorithmic traders. Current research investigates efficient economic mechanisms that fully automate the provisioning and usage processes of Gridbased services. On the one hand, consumers want to allocate resources on demand for their various applications, e.g. data sharing, stream processing, email, business applications and simulations. On the other hand, providers of Grid services want to govern business policies to meet their utilization and profit goals. The above-mentioned processes are not manually manageable, however, because decisions need to be taken within milliseconds. Therefore, such processes have to be automated to minimize human interactions. Hence, market mechanisms and strategic behavior play important roles when it comes to achieving automated and efficient allocation of Grid services. The paper begins by presenting a framework for automated bidding, providing a methodology for the design and implementation of configurable bidding strategies. Second, it presents a novel bidding strategy based on a reinforcement learning technique. This strategy is designed to automate the bid generation processes of consumers and providers in various market mechanisms. Third, the behavior and convergence of the strategy is evaluated in a centralized Continuous Double Auction and a decentralized on-line machine scheduling mechanism against selected benchmark bidding strategies.

Market-oriented approach is an effective method for resource management because of its regulation of supply and demand and is suitable for cloud environment where the computing resources, either software or hardware, are virtualized and allocated as services from providers to users. In this paper a continuous double auction method for efficient cloud service allocation is presented in which i) enables consumers to order various resources (services) for workflows and coallocation, ii) consumers and providers make bid and request prices based on deadline and workload time and in addition providers can tradeoff between utilization time and price of bids, iii) auctioneers can intelligently find optimum matching by sharing and merging resources which result more trades. Experimental results show that proposed method is efficient in terms of successful allocation rate and resource utilization.