Olivier Brandouy’s research while affiliated with University of Lille Nord de France and other places

The purpose of this paper is to define software engineering abstractions that provide a generic framework for stock market simulations. We demonstrate a series of key points and principles that has governed the development of an Agent-Based financial market application programming interface (API). The simulator architecture is presented. During artificial market construction we have faced the whole variety of agent-based modelling issues : local interaction, distributed knowledge and resources, heterogeneous environments, agents autonomy, artificial intelligence, speech acts, discrete or continuous scheduling and simulation. Our study demonstrates that the choices made for agent-based modelling in this context deeply impact the resulting market dynamics and proposes a series of advances regarding the main limits the existing platforms actually meet.

We study the impact of the network topology on various market parameters (volatility, liquidity and efficiency) when three populations or artificial trades interact (Noise, Informed and Social Traders). We show, using an agent-based set of simulations that choosing a Regular, a Erdös-Rényi or a scale free network and locating on each vertex one Noise, Informed or Social Trader, substantially modifies the dynamics of the market. The overall level of volatility, the liquidity and the resulting efficiency are impacted by this initial choice in various ways which also depends upon the proportion of Informed vs. Noise Traders.

We study the impact of the network topology on various market parameters (volatility, liquidity and efficiency) when three populations or artificial trades interact (Noise, Informed and Social Traders). We show, using an agent-based set of simulations that choosing a Regular, a Erdös-Rényi or a scale free network and locating on each vertex one Noise, Informed or Social Trader, substantially modifies the dynamics of the market. The overall level of volatility, the liquidity and the resulting efficiency are impacted by this initial choice in various ways which also depends upon the proportion of Informed vs. Noise Traders.

Randomness and regularities in Finance are usually treated in probabilistic terms. In this paper, we develop a completely different approach in using a non-probabilistic framework based on the algorithmic information theory initially developed by Kolmogorov (1965). We present some elements of this theory and show why it is particularly relevant to Finance, and potentially to other sub-fields of Economics as well. We develop a generic method to estimate the Kolmogorov complexity of numeric series. This approach is based on an iterative "regularity erasing procedure" implemented to use lossless compression algorithms on financial data. Examples are provided with both simulated and real-world financial time series. The contributions of this article are twofold. The first one is methodological : we show that some structural regularities, invisible with classical statistical tests, can be detected by this algorithmic method. The second one consists in illustrations on the daily Dow-Jones Index suggesting that beyond several well-known regularities, hidden structure may in this index remain to be identified.

We explore the potential benefits of a series of existing and new non-parametric convex and non-convex frontier-based fund rating models to summarize the information contained in the moments of the mutual fund price series. Limiting ourselves to the traditional mean-variance portfolio setting, we test in a simple backtesting setup whether these efficiency measures fare any better than more traditional financial performance measures in selecting promising investment opportunities. The evidence points to a remarkable superior performance of these frontier models compared to most, but not all traditional financial performance measures.

We survey the main applications of algorithmic (Kolmogorov) complexity to the problem of price dynamics in financial markets. We stress the differences between these works and put forward a general algorithmic framework in order to highlight its potential for financial data analysis. This framework is “general" in the sense that it is not constructed on the common assumption that price variations are predominantly stochastic in nature.

We study the Volume Synchronized Probability of Informed Trading (VPIN) proposed by Easley D, López de Prado M, O’Hara (Rev Financ Stud 25:1457–1493, 2010) as a consistent measure of the “order flowtoxicity”. The VPIN is a proxy for the probability that informed traders adversely select uninformed ones, notably Market Makers.We use a price-driven, asynchronous, agent-based artificial market where populations of agents evolve according to the general logic and within a similar framework as proposed by Easley D, Kiefer D, O’Hara M, Paperman J (J Financ 51(4):1405–1436, 1996). Among others, we document situations in which the VPIN is at high levels even if no informed trading is at play. This ambiguity in the consistency of the VPIN suggests that this measure may mislead competitive market makers in their decisions about the spread.

Convex and nonconvex frontier rating models have been in use for some time to assess the relative efficiency of mutual funds (MF) with regard to the costs incurred by investors and the moment characteristics of the return series generated over time. The rating models in this contribution use the shortage function as an inefficiency measure for each MF relative to the investment universe studied. This approach reveals a set of efficient MF that may be ideal candidates for an investor willing to construct a superfund. This chapter offers the first detailed backtesting analysis of these frontier-based MF ratings compared to a series of more traditional financial performance measures as well as one professional rating (in casu, Morningstar). In combination with various diversification procedures (from naïve equally weighted portfolios to sophisticated multi-moment optimization), we show that strategies using frontier-based ratings perform very well compared to strategies founded on more traditional rating techniques. These strategies prove to be among the best performing strategies analyzed under the various test scenarios (e.g., a variety of sample sizes, etc.). Therefore, frontier-based ratings offer promising perspectives for active investors and may well supply new quantitative tools for superfund managers.

This paper proposes a new method for determining the upper bound of any investment strategy's maximum profit, applied in a given time window [0,T]. This upper bound is defined once all the prices are known at time T and therefore represents the ex-post maximum efficiency of any investment strategy determined during the relevant time interval. This approach allows us to gauge in absolute terms those behaviors defined through atomic “buy” and “sell” actions, and can be extended to more complex strategies. We show that, even in the ex-post framework, establishing this upper bound when transaction costs are implemented is extremely complex. We first describe this problem using a linear programming framework. Thereafter, we propose to embed this question in a graph theory framework and to show that determining the best investment behavior is equivalent to identifying an optimal path in an oriented, weighted, bipartite network or a weighted, directed, acyclic graph. We illustrate this method using real world data and introduce a new theory about absolute optimal behavior in the financial world.