Serafín Martínez-Jaramillo

An Heterogeneous, Endogenous and Coevolutionary GP-Based Financial Market

Stock markets are very important in modern societies and their behavior has serious implications for a wide spectrum of the worlds population. Investors, governing bodies, and society as a whole could benefit from better understanding of the behavior of stock markets. The traditional approach to analyzing such systems is the use of analytical models. However, the complexity of financial markets represents a big challenge to the analytical approach. Most analytical models make simplifying assumptions, such as perfect rationality and homogeneous investors, which threaten the validity of their results. This motivates alternative methods.In this paper, we report an artificial financial market and its use in studying the behavior of stock markets. This is an endogenous market, with which we model technical, fundamental, and noise traders. Nevertheless, our primary focus is on the technical traders, which are sophisticated genetic programming based agents that co- evolve (by learning based on their fitness function) by predicting investment opportunities in the market using technical analysis as the main tool. With this endogenous artificial market, we identify the conditions under which the statistical properties of price series in the artificial market resemble some of the properties of real financial markets. By performing a careful exploration of the most important aspects of our simulation model, we determine the way in which the factors of such a model affect the endogenously generated price. Additionally, we model the pressure to beat the market by a behavioral constraint imposed on the agents reflecting the Red Queen principle in evolution. We have demonstrated how evolutionary computation could play a key role in studying stock markets, mainly as a suitable model for economic learning on an agent- based simulation.

Financial contagion: extending the exposures network of the Mexican financial system

Direct contagion has been widely studied in recent years and little evidence has been found to be relevant to the study of systemic risk. However, we argue that this limited contagion effect might be associated with a lack of relevant data. A common assumption for the estimation of the matrices of exposures is to apply the maximum entropy principle to deal with data gaps; such an assumption might lead to an underestimation of contagion risk. In this paper, there are no data gaps and the information set is extended from interbank exposures alone to exposures among most of the financial intermediaries in the Mexican financial system (we even include exposures to some international foreign banks). Naturally, the contagion risk of an extended network of exposures changes with respect to the interbank exposures network, as there are many more institutions which can be the source of contagion and there are more institutions which can fail due to contagion. The most important contribution of this paper is that it provides evidence on financial contagion with an extended exposures network under stressful conditions. The results presented here support the international efforts by the Bank for International Settlements, the International Monetary Fund and the Financial Stability Board to increase the amount of information available which can be used to assess systemic risk and contagion based on exposures and funding data.

A multiplex network analysis of the Mexican banking system: link persistence, overlap, and waiting times

This paper analyzes the persistence and overlap of relationships between banks in a multiplex decomposition of the exposures network. Our analysis may be useful for researchers designing stress tests or models in which the behavior of banks is modeled explicitly. This has not been looked at previously, considering the time period involved and the different types of exposures and interactions used. We show that trading relationships overlap for some pairs of banks, and link persistence is higher in the secured than the unsecured market. Moreover, link persistence in the securities cross-holding network is much higher than in other funding networks, and overlap with the other segments of activity is low, despite being persistent over time. Additionally, unsecured loans received by large banks have the shortest waiting times (that is, for a given borrower and a given lender, the number of days elapsed before a new loan is observed) regardless of counterparty size, which suggests quicker access to liquidity. Large banks lend (unsecured) with shorter waiting times to medium-sized banks than to small banks; this is not the case when they lend in the secured layer of the network. Small banks have quicker access to liquidity in the secured lending layer when borrowing from medium-sized banks.

Simulation in Computational Finance and Economics Tools and Emerging Applications

The works presented in this book can be used as an inspiration for economic researchers interested in creating their own computational models in their respective fields.

Understanding Bank Failure: A Close Examination of Rules Created by Genetic Programming

This paper presents a novel method to predict bankruptcy, using a Genetic Programming (GP) based approach called Evolving Decision Rules (EDR). In order to obtain the optimum parameters of the classifying mechanism, we use a data set, obtained from the US Federal Deposit Insurance Corporation (FDIC). The set consists of limited financial institutions’ data, presented as variables widely used to detect bank failure. The outcome is a set of comprehensible decision rules, which allows to identify cases of bankruptcy. Further, the reliability of those rules is measured in terms of the true and false positive rate, calculated over the whole data set and plot over the Receiving Operating Characteristic (ROC) space. In order to test the accuracy performance of the mechanism, we elaborate two experiments: the first, aimed to test the degree of the variables’ usefulness, provides a quantitative and a qualitative analysis. The second experiment completed over 1000 different re-sampled cases is used to measure the performance of the approach. To our knowledge this is the first computational technique in this field able to give useful insights of the method’s predictive structure. The main contributions of this work are three: first, we want to bring to the arena of bankruptcy prediction a competitive novel method which in pure performance terms is comparable to state of the art methods recently proposed in similar works, second, this method provides the additional advantage of transparency as the generated rules are fully interpretable in terms of simple financial ratios, third and final, the proposed method includes cutting edge techniques to handle highly unbalanced samples, something that is very common in bankruptcy applications.

Bankruptcy Prediction for Banks: An Artificial Intelligence Approach to Improve Understandability

Artificial Intelligence (AI) is a prominent field within Computer Science whose main goal is automatic problem solving. Some of the foundations of this area were established by Alan M. Turing in his two seminal papers about machine intelligence [39] and [40]. Machine Learning (ML) is an important branch within the AI field which currently is on an intensive stage of development due to its wide range of applications. In particular, ML techniques have recently gained recognition in finance, since they are capable to produce useful models. However, the difficulty, and even the impossibility, to interpret these models, has limited the use of ML techniques in some problems where the interpretability is an important issue. Bankruptcy prediction for banks is a task which demands understandability of the solution. Furthermore, the analysis of the features (input variables), to create prediction models, provides better knowledge about the conditions which may trigger bank defaults. The selection of meaningful features before executing the learning process is beneficial since it reduces the dimensionality of the data by decreasing the size of the hypothesis space. As a result, a compact representation is obtained which is easier to interpret. The main contributions of this work are: first, the use of the evolutionary technique called Multi-Population Evolving Decision Rules MP-EDR to determine the relevance of some features from Federal Deposit Insurance Corporation (FDIC) data to predict bank bankruptcy. The second contribution is the representation of the features’ relevance by means of a network which has been built by using the rules and conditions produced by MP-EDR. Such representation is useful to disentangle the relationships between features in the model, this representation is aided by metrics which are used to measure the relevance of such features.

Evolutionary Computation and Artificial Financial Markets

It is essential not only for investors but for regulators to understand the mechanisms that govern financial markets. However, financial markets are constantly evolving and are becoming more complex and as a consequence more difficult to analyze and understand. Traditional analytical methods cannot explain some of the phenomena which are present in real markets and some of the assumptions that had to be made for the sake of tractability in such models are over-simplistic. This opens the field to alternative methods that allow us to relax some of the most unrealistic assumptions in order to gain a better understanding of such complex systems. Agent-based computational economics (ACE) offers a suitable alternative for the study of financial markets. In this chapter we develop a software platform called Co-evolutionary, Heterogeneous Artificial Stock Market (CHASM); which allows us to perform a series of experiments with the purpose of identifying the aspects that could be responsible for the statistical properties (stylized facts) of financial prices. In CHASM, we model different types of traders: technical, fundamental and noise traders. However, we focus our research on technical traders represented as genetic programming (GP) based agents which co-evolve in the market forecasting price changes on the basis of technical indicators. We perform a detailed exploration of the market’s features in order to identify the conditions under which the stylized facts emerge. Moreover, we develop a behavioral constraint inspired by the Red Queen evolutionary principle to model endogenously the competitive pressure of the market.

The repo market in Mexico: Empirics and stylized facts☆

Abstract This paper is a thorough description of the repo market in Mexico, a funding market of significant importance for Mexican commercial banks, brokerage houses, and development banks, and which, unlike the repo markets in Europe and the U.S., is an OTC market with no central counterparty or tri-partite repos. In this paper, we describe the counterparties which are involved in repo transactions, as well as distribution by collateral type, and provide some further metrics on the trading volume and other informative statistics on haircuts and rates. Given that Banco de Mexico possesses information on individual repo transactions from 1998, we also study the evolution of this market, paying particular attention to the financial crisis that started in 2007. This is one of a few descriptive papers on this market due to the extremely detailed and granular data base on repo transactions in Mexico. The main contribution of this paper is to document and describe the structure of a market for which there exists a long history of individual repo transactions and which is of great importance from the perspective of the funding structure of banks.