Search

Artificial multi-agent systems can be applied to replicate complex processes, including the stock market. In this manuscript, an artificial stock market was simulated. Heterogeneous agents are set to trade by using different investiment strategies through a double auction market. The main innovative aspect of this work is to evaluate the performance and effectiveness of agents with different investment strategies based on the combination of Technical/Fundamental Analysis. According to the simulations performed, the results showed that the time series resulting from the interactions among these agents resemble to the real financial series, reproducing some of the stylized facts found in the Literature. Moreover, the statistical analysis reported no significant differences in the distribution mean performance between these diverse agents types.

A new trading system (TS) is proposed in this paper, based on singular spectrum analysis (SSA) and on a decision-making process that relies on performance measures adjusted by profit and risk. The aim is to contribute to the literature of technical analysis (TA) by proposing a general and adaptive tool for supporting trading strategies of investors and market practitioners. Recent advances of trading rules based on SSA are pooled together in a complimentary fashion, with the intention that this group of strategies outperforms its individual constituents, as occur with ensembles. Simulations are presented showing the improvements of the proposed TS over classic tools of TA and the benchmark Buy & Hold strategy for the Brazilian stock market.

The stock price movements result from many factors that are often difficult to be detected and modelled. The investigation of price trends and the use of the information available to evaluate investments and identify trading opportunities can be promising. However, financial data are non-stationary, i.e., their statistical characteristics constantly change. Therefore, the financial market is a challenging environment for the application of Machine Learning techniques, since they can only make reliable predictions to data consistent with what they have seen before. This paper test the use of a Machine Learning technique known as Support Vector Machines (SVM) aiming at being a tool to support the decision making process of trading at the stock market. SVM aggregates some input signals and, based on a set of technical indicators and historical price changes, create buy/sell recommendations of a given security as outputs. The dataset comprises several Brazilian stocks time-series traded on both the Brazilian (B3) and American (NYSE) stock exchange. These time-series belong to various economic sectors and present different market dynamics. The computational simulations are based on a fictitious strategy that does not consider the trading costs and only long positions are allowed. Using two risk-adjusted performance metrics, the results show that strategies based on the SVM model achieve better performance than the Buy & Hold benchmark.

Singular Spectrum Analysis (SSA) is a non parametric approach that can be used to decompose a time series into trends, oscillations and noise. In online applications, the SSA algorithm must be recalculated for each new sample available. The so called Causal SSA have been used in this context. This paper aims to contribute to the application of Causal SSA as an indicator for trading rules design by means of comparing them to previously proposed ones based on classic indicators. Using historical time series from some of the major Brazilian stocks, different simulations are carried out. Some performance metrics are applied along with several statistical tests, thereby supporting a detailed analysis of the experimental results obtained. The results analysis indicates that, no matter what technical trading rules, it is worth to apply the SSA method rather than classic approaches.

The Singular Spectrum Analysis (SSA) is powerful method, capable of working with arbitrary statistical process and it is adaptive to the underlaying data. Many variations of the standard methodology have been prosed in recent years improving the performance, adjusting to specific problems or objectives, or addressing some shortcomings. One of such drawbacks occurs when the spectrum spreads and varies over time, demanding many elementary matrices to reconstruct an approximation of the original series, hampering the method applicability. Also, another difficulty arises when large datasets are analyzed. There are computational issues and also problems with the method ability to maintain satisfactory separability. To circumvent these issues, a new method has been proposed. The original time series is divided into smaller and consecutive segments, with some superposition between them. Then, standard SSA is applied to each segment and the results are concatenated properly. This paper provides an implementation of this algorithm and some experiments are shown to illustrate the improvements achieved. Keywords: Singular spectrum analysis, Non-stationary signals, Segmentation

Singular Spectrum Analysis (SSA) is a powerful non-parametric framework to analysis and enhancement of time-series. SSA may be capable of decomposing a time-series into its meaningful components: trends, oscillations and noise. However, if the signal under analysis is non-stationary, with its spectrum spreading and varying in time, the reliability of the reconstruction is guaranteed only when many elementary matrices are used. As a consequence, the capability to discriminate dominant structures from time-series may be impaired. To circumvent this issue, a new method, called overlap-SSA (ov-SSA), is proposed for segmentation, analysis and reconstruction of long-term and/or non-stationary signals. The raw time series is divided into smaller, consecutive and overlapping segments, and standard SSA procedures are applied to each segment with the resulting series being concatenated. This variation of SSA seeks to: improve reconstruction and component separability for non-stationary time-series; enable the analysis for large datasets, avoiding the issues of concatenation of many segments; and present some benefits of the segmentation in terms of better time–frequency characterization. These advantages are illustrated in several synthetic and experimental datasets.

A large effort has been made for improving the quality of electric power in the last years. More often, the studies concern over methods and techniques to enhance monitoring systems. One of the major relevance is the fundamental frequency estimation, especially in the presence of harmonics and interharmonics disturbances. This paper focuses on the accurate frequency estimation of power signals corrupted by stationary white noise and harmonics and interharmonics disturbances using the least mean square (LMS) method, a simple and well-known technique. The effects of these disturbances are studied in an extensive set of simulation results which demonstrate the feasibility of the LMS method to estimate the fundamental frequency of corrupted signals.

The advent of Internet of Things - IoT - technology has elevated the embedded solutions employment to level never seen before, both in the problem-solving possibilities as in the complexity of the new systems. SoC-based computers such as those used in embedded systems have been under growing attention due to their low-cost, low-power, reduced footprint and high performance characteristics. These features allow small users as well as industries to design, develop and build wide range of applications with efficiency as the common aspect. A number of platforms that integrate this kind of technology were also proposed by different manufacturers, some employing the open hardware concept. Embedded applications spread over a wide range of fields, which make them hard to characterize in a general sense. For this reason, benchmark suites are often used as a tool for system’s performance evaluation. However, such suites consist on a large number of different algorithms, which can spend time and energy to run on SoC computers. In order to contribute to the performance assessment of such systems, this paper presents a simple method for comparing different platforms, based on execution time of simple but representative algorithms: sorting, cross-correlation and discrete-cosine transform calculation for vectors. The power consumption was also considered. Six distinct architectures were used as a case study. Several computer experiments were conducted in order to make possible a statistic characterization of different load conditions in distinct platforms, so their comparison can be done in a meaningful way. Results show that the proposed method provides a good estimation of relative system performance, which can be useful as a criterion choice for selecting different embedded platforms.

The objective of this work is to analyze the productivity of different Brazilian research promotion agencies by using the Method of Interactive Decision Making and Multicriteria (Acronym TODIM in Portuguese), which is based on the Prospect Theory. The proposed model classify of 22 Brazilian research agencies (two federal and twenty state agencies) under six different criteria. The result shows that some small state agencies are more efficient than bigger ones in terms of the funding/scholarship ratio. This suggests that the winner model could be further investigated to be promoted among the other agencies of this country.

Singular Spectrum Analysis (SSA) is a nonparametric approach that can be used to decompose a time series as trends, oscillations and noise. In online applications, the SSA algorithm must be recalculated for each new sample available. The so called Causal SSA have been used in this context. In this manuscript, an alternative version of online SSA, the Recursive SSA, is proposed as a technical indicator. Based on a “forgetting factor” parameter, it is possible to control the amount of previous samples that are used in the SSA algorithm. This functionality may confer adaptive and hybrid features, thereby providing a crucial characteristic to technical indicators. The Recursive SSA technical trading rules (SSA-TTR) are applied to the DJIA time-series and compared against popular technical indicators. The results show the advantages of SSA-TTR over the popular ones.

This manuscript introduces a Matlab computational framework for a multiagent system simulation of financial markets. The main motivation of using Matlab is that this platform is very familiar among researches in the area of engineering. This could motivate that researchers to investigate this area. The aim of our platform is to replicate the interactions among agents through a double auction environment, leading to an artificial market. Every agent is assigned with unlimited credit, and short selling is allowed. These two degrees of freedom might ensure that a particular agent is able to trade whenever it becomes active. Agents should be able to take their own decisions. The key innovative aspect of this work is the development of a computational platform which integrates such heterogeneous financial agents. This framework intends to mimic behavioral patterns followed by the market practitioners. Even though the proposed framework uses a simplified version of the financial market, the time series resulted from the interactions of those agents reproduces some of the financial stylized facts. Therefore, at least some market dynamics are successfully emulated by the proposed approach.

This paper describes the development of an autonomous navigation system for a small-scale ground vehicle which uses four independent DC motors for propulsion and an Arduino Mega2560 as its onboard computing plataform. The vehicle sensory system is composed of a low-cost Inertial Measurement Unit (which includes 3 magnetometers) and a standard GPS receiver. The onboard computer sends data from the onboard sensors to a Ground Control Station (GCS) and receives navigation commands from it. These commands are then used to drive the ground vehicle’s DC motors using PWM (Pulse-Width Modulation) signals. The Ground Control Station software implements an Extended Kalman Filter for sensor data fusion. An ethernet shield and an ethernet wifi bridge are used for wireless communication between the vehicle onboard computer and the Ground Control Station. A two-wheeled vehicle model was used to develop the vehicle controller and to tune the Kalman Filter parameters. Simulations and real-world test results of the proposed autonomous navigation procedure are presented and discussed for a path-following experiment.

This work intends to replicate the interactions among intelligent agents in a double auction environment, leading to an artificial market. Aiming at making agents able to take their own decisions, different methodologies for agents intelligence are proposed, ranging from zero intelligence agents to reinforcement learning. Regarding the study of financial markets, agents can rely on a myriad of techniques in their making-decision process. In this manuscript, the Fundamental and/or Technical Analysis are the tools used by the agents to take their own decisions. A different set of parameters is assigned to each agent, thereby heterogeneity is included within each strategy. Based on these simple strategies some agents can also learn with their past experience, updating the probability to select a trading strategy according to switching rules as a function of the previous result. Several experiments are carried out. Results show that the time series produced by the interactions of those agents resembles to financial time series, reproducing some stylized facts encountered thereof. The performance of different methods for agent design are also compared.

A comparison between Moving Averages (MA) and two versions of Singular Spectrum Analysis (SSA) methodology — the Caterpillar and the Toeplitz — is presented. Caterpillar had already been studied in this manner but the same is not true for the Toeplitz SSA. Toeplitz SSA assumes the stationarity of the time-series, which means that the process needs to be mean-reverting. However, such assumption is not a necessary condition for the Caterpillar SSA. In this paper both approaches are applied to a trend estimation problem in order to be used as an indicator in trend-following technical rules design. Similarities and differences between these techniques are addressed. The obtained results suggest that, although SSA approaches provides more flexibility to achieve a desired trend resolution compared to the traditional MA, the Toeplitz SSA exhibit some issues that might put it off its use in this particular application.

In this paper, we introduce a novel method to select the components of a portfolio of securities. This method is based on a reinforcement learning technique known as learning automata. Several heuristic solutions for the portfolio weights selection problem have been introduced in literature. The point is that these applications assumes that portfolio components are given. The difference of our work is that we propose some heuristic in order to select the portfolio components instead of the weights. In terms of heuristic, we propose learning automata because its ability to solve complex systems such as a the optimal portfolio components. We test the use of learning automata in terms of financial indicators optimization. Our findings show that our proposed method improves the portfolio optimization performance in terms of accuracy and computational effort.

Neste trabalho ´e apresentado um sistema para opera¸c˜ao de manipuladores rob´oticos baseado em t´ecnicas de Vis˜ao Computacional. As imagens, representativas de gestos prestabelecidos, s˜ao capturadas por meio do sensor do tipo RGB-D e processadas por uma t´ecnica de Aprendizado de M´aquina, conhecida comoM´aquina de Vetores de Suporte (Support Vector Machines ou SVM). Uma vez classificados, cada gesto indica um movimento diferente a ser executado pelo manipulador rob´otico. Os comandos s˜ao enviados para esse dispositivo via comunica¸c˜ao serial. Os resultados obtidos demonstram que foi poss´ivel a captura, processamento, classifica¸c˜ao e envio das informa¸c˜oes corretas ao manipulador para execu¸c˜ao dos movimentos predeterminados.

The connection between Singular Spectrum Analysis (SSA) decomposition and short-term market movements is investigated. Since SSA is a non-parametric approach, suitable to decompose general time-series into meaningful components, such as trends, oscillations and noise, it is proposed as a new oscillator-type Technical Indicator, replacing popular ones. New Technical Trading Rules (TTRs) are designed and applied to some major global stock indexes to illustrate the benefits in terms of revealing market movements. The performance is evaluated according to different risk-adjustment metrics and the empirical results reveal that the SSA-TTRs may outperform some popular technical oscillators and also the Buy & Hold strategy.

This paper proposes an optimization-based approach to the robust H∞ control problem of uncertain continuous or discrete-time linear time-invariant systems with different time-varying delays in the state vector and control input of the dynamic equation and controlled output. Sufficient delay-dependent conditions are derived for the control stabilization problem, where both the size of the time-varying delay and the size of its time derivative (in the continuous-time case) play a crucial role for the closed-loop stability with a guaranteed H∞ performance index. The solutions that are proposed for the H∞ control problem are similar or less conservative, when compared to other recent approaches.

A linear matrix inequality approach to the robust H∞ control problem of uncertain continuous- and discrete-time linear time-invariant systems with time delay in the state vector and control input is developed. The main results provide sufficient delay-dependent conditions for the control problem, where the explicit size of the time delay plays a crucial role for the closed-loop stability. The solutions that are found for the H∞ control problem are less conservative when compared with other approaches in the recent literature.

This paper proposes a Linear Matrix Inequality (LMI) approach to the robust H ∞ control stabilization problem of uncertain continuous- or discrete-time linear time-invariant systems with time-varying delays in the state vector and control input. It is derived sufficient delay-dependent conditions for the control stabilization problem, where both the size of the time-varying delay and the size of its time derivative (applicably only in the continuous-time case) play a crucial role for the closed-loop stability with a guaranteed H ∞ level. The solutions that are found for the H ∞ control problem are similar or less conservative, when compared to others approaches.

Singular Spectrum Analysis (SSA) is a non-parametric approach that can be used to decompose a time-series as trends, oscillations and noise. Trend-following strategies rely on the principle that financial markets move in trends for an extended period of time. Moving Averages (MAs) are the standard indicator to design such strategies. In this study, SSA is used as an alternative method to enhance trend resolution in comparison with the traditional MA. New trading rules using SSA as indicator are proposed. This paper shows that for the Down Jones Industrial Average (DJIA) and Shangai Securities Composite Index (SSCI) time-series the SSA trading rules provided, in general, better results in comparison to MA trading rules.

This paper develops a Linear Matrix Inequality (LMI) approach to the robust /spl Hscr//sub /spl infin// control problem of uncertain continuous-and discrete-time linear time-invariant systems with time-delay in the state vector and control input. The main results provide sufficient delay-dependent conditions for the control problem, where the explicit size of the time delay plays a crucial role for the closed-loop stability. The solutions that are found for the /spl Hscr/;/sub /spl infin// control problem are less conservative, when compared to other approaches.