No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
The role of genetic algorithms and wavelets in computational intelligence based decision support for stock exchange daily trading
... The selection technique was the stochastic sampling with replacement (SSR). The fitness values were rescaled after each iteration in order to avoid premature convergence.The heuristic variable point crossover 6 scheme presented in previous work [30] [31] was adopted here too, while it was proved more efficient than any fixed point scheme. The genetic procedure tracks the optimum chromosome throughout the operation and when the desirable criteria are met it adapts this chromosome as the best rule base. ...
... The selection technique was the stochastic sampling with replacement (SSR). The fitness values were rescaled after each iteration in order to avoid premature convergence.The heuristic variable point crossover 6 scheme presented in previous work [30, 31] was adopted here too, while it was proved more efficient than any fixed point scheme. The genetic procedure tracks the optimum chromosome throughout the operation and when the desirable criteria are met it adapts this chromosome as the best rule base. ...
... The results show that the proposed system has better performance than three other wavelet thresholding algorithms (cross-validation, best basis selection and best level tree). In [116] and [117], the authors propose a multistage hybrid system combining wavelet thresholding, neural networks and neuro-fuzzy systems for stock exchange daily trading. The system is proved to be superior to individual components performance and to buy and hold strategies. ...
... The results show that the proposed system has better performance than three other wavelet thresholding algorithms (cross-validation, best basis selection and best level tree). In [116] and [117], the authors propose a multistage hybrid system combining wavelet thresholding, neural networks and neuro-fuzzy systems for stock exchange daily trading. The system is proved to be superior to individual components performance and to buy and hold strategies. ...
The increased popularity of hybrid intelligent systems in recent times lies to the extensive success of these systems in many real-world complex problems. The main reason for this success seems to be the synergy derived by the computational intelligent components, such as machine learning, fuzzy logic, neural networks and genetic algorithms. Each of these methodologies provides hybrid systems with complementary reasoning and searching methods that allow the use of domain knowledge and empirical data to solve complex problems. In this paper, we briefly present most of those computational intelligent combinations focusing in the development of intelligent systems for the handling of problems in real-world applications.
... The selection technique was the stochastic sampling with replacement (SSR). The fitness values were rescaled after each iteration in order to avoid premature convergence.The heuristic variable point crossover 10 scheme presented in previous work [Tsakonas et al. 2000] was adopted here too, while it was proved more efficient than any fixed point scheme. An interpetation for this result may involve the similarity which two subsequent outputs may have in our system, in terms of neighboring in the derivation of (an ordered) rule base. ...
: This paper reflects our study on the efficiency and the characteristics of a fuzzy rulebased system when used for forecasting in noisy data such as stock market returns for daily decision making. From previous work, we may consider that, such a kind of data is usually distinguished by a stochastic component, and an underlying trend which, exhibits nonlinear dynamic relationship between subsequent values. Based on recent demonstrations showing that, local fuzzy reconstruction methods can identify and predict such complex dynamic processes, we practice the implementation of a fuzzy rule-based system directly on this data. Our main concerns involve both the selection and evaluation of training data as well as the system configuration and the learning procedure itself. The latter includes the self-learning of the rule base by evolving procedures such as genetic algorithms and the membership function fine tuning using neuro-fuzzy techniques. In order to avoid the error amplification during the subsequent prediction, as a result of sensitivity to initial conditions, a statistical error monitoring procedure is followed. For the evaluation of this system, various simulation strategies can be applied which demonstrate that the system is capable of short-term prediction in realworld noisy nonlinear dynamic processes, offering an almost safe trading policy in terms of cumulative return over a given time horizon. KEYWORDS: Genetic algorithms, neuro-fuzzy systems, time-series forecasting, fractal dimension, financial decision support. 1.
ResearchGate has not been able to resolve any references for this publication.