Introduction to Expert Systems, Fuzzy Logic, Neural Networks, and Chaos Theory

The power of genetic algorithms (GAs) and related expert systems such as fuzzy logic, neural networks, and chaos theory and other classifier systems is truly infinite in nature. The above stated procedures are sure to happen in the near future, and there is no chance for it not to occur. GAs, fuzzy logic, neural networks, and chaos theory are all biologically-inspired algorithmic procedures, as they all are linked to the world of biology in some way. Market represents the ideas of traders. In the present environment, the market is driven by the ideas generated by the use of these AI-based expert systems and it is causing huge competition in making profits. This chapter is planned to be a detailed introduction of various popular expert systems such as GAs, neural networks, fuzzy logic, and chaos theory and their usages. Researchers in the past have proved that these computational procedures could have far reaching effects in the stock trading system.

Using Genetic Algorithms to Develop Investment Strategies

Genetic algorithms (GAs) are a powerful search technique. The use of genetic algorithms (GAs) will help in the development of better trading systems. The genetic algorithms (GAs) help the researcher to explore various combinations of trading rules or their parameters, which the human mind is unable to find. This chapter explains genetic algorithms (GAs) in brief and gives insight on how they find better trading strategies. Some of the manual trading strategies are good in nature. Genetic algorithms (GAs) only addition to them. Interfacing genetic algorithms (GAs) with stock trading systems or developing a combined model requires a large degree of imagination and creativity. It is an art not a scientific invention. Genetic algorithms (GAs) make use of computers to find various interesting trading systems.

Operator Control Parameters and Fine Tuning of Genetic Algorithms (GAs)

Genetic algorithms (GAs) are heuristic, blind (i.e., black box-based) search techniques. The internal working of GAs is complex and is opaque for the general practitioner. GAs are a set of interconnected procedures that consist of complex interconnected activity among parameters. When a naive GA practitioner tries to implement GA code, the first question that comes into the mind is what are the value of GA control parameters (i.e., various operators such as crossover probability, mutation probability, population size, number of generations, etc. will be set to run a GA code)? This chapter clears all the complexities about the internal interconnected working of GA control parameters. GA can have many variations in its implementation (i.e., mutation alone-based GA, crossover alone-based GA, GA with combination of mutation and crossover, etc.). In this chapter, the authors discuss how variation in GA control parameter settings affects the solution quality.

Genetic Algorithms (GAs) and Their Mathematical Foundations

In this chapter, the authors back GA procedures using old mathematical facts. More rigorous working of mathematical facts about GAs are raised in this chapter. In fact, there are a large number of similarities in the population of strings. The authors see how GA exploits these similarities to generate good solutions. So, in this whole procedure they show which schema or pattern will grow and which pattern will die or be lost as generation passes by due to the effect of selection, crossover, and mutation operator. The study of this building block hypothesis, leads to better understanding of GA. It will also help us to reach optimal solutions in much less time.

Developing a Single Indicator or Multiple Indicator Market Timing System

In this chapter, the authors use genetic algorithms (GAs) to optimize the parameters of the trading system, which is made by various technical indicators. These trading systems or rules will give buy or sell signals when applied on past prices of a particular stock. Genetic algorithms (GAs) have an ability to find optimal trading indicators that will predict the market direction or trend with greater accuracy. Use of genetic algorithms (GAs) in conjunction to a trading rule refutes efficient market hypothesis (EMH) in a weak form.

Introduction to Biologically Inspired Algorithms

In order to find more sophisticated ways to remain in competition in the stock market, investors and analysts are finding procedures based on nature-inspired artificial intelligence-based algorithms. It is seen that interest of researchers has grown in these technologies in the past years. These newer techniques have changed the investment arena of the stock market. A lot of thought process, hard work, creativeness, and knowledge about these algorithms are required to implement them in the stock investment area. In the past, few people have had the privilege to implement and obtain better results by using these algorithms. But with the access to affordable computing systems and experts with the knowledge of these computing systems, we can take advantage of making profit from the market. This chapter explains the detail working of these AI techniques such as chaos theory, neural networks, fuzzy logic, and genetic algorithms in detail.

Genetic Algorithm (GA) Methodology and Its Internal Working

Many practitioners are shy with implementing GAs. Due to this, a lot of researchers avoid using GAs as problem-solving techniques. It is desirable that an implementer of GA must be familiar in working with high-level computer languages. Implementation of GA involves complex coding and intricate computations which are of a repetitive nature. GAs if not implemented with caution will result in vague or bad solutions. This chapter overcomes the obstacles by implementing and defining various data structures required for implementing a simple GA. They will write various functions of GA code in C ++ programming language. In this chapter, initial string population generation, selection, crossover, and mutation operator used to optimize a simple function (one variable function) coded as unsigned binary integer is implemented using C ++ programming language. Mapping of fitness issue is also discussed in application of GAs.

Synergistic Market Analysis, Technical Analysis, and Various Indicators

Most of the stock and financial market analysis uses past or historic data in order to forecast the future market indices. This study of past or historic data in order to infer certain value addition from it is known as technical analysis. In this chapter, the authors study a large number of popular indicators. A trader uses these indicators individually or in combination with other indicators to make a trading rule. They test them on past data and choose particular parameters and indicators that gave more profit and drop those that are loss making in nature. Thus, the decision to make entry and exit is given by these technical indicators. This chapter gives a detailed explanation of the most popular technical indicators in use.

Some Other Applications of Genetic Algorithms (GAs)

As we had already seen that genetic algorithms (GAs) are smart in their working. Here, the authors explore the rich working of genetic algorithms (GAs) in various diversified fields. Until now, they had discussed the historical nature of genetic algorithms (GAs). They have also discussed the programming code to run simple genetic algorithms (SGA). Lastly, they are going to take an overview of the application of genetic algorithms (GAs) in various fields (i.e., from business to non-business). Already, they have discussed the robust working of genetic algorithms (GAs) in various adverse conditions. Here, they discuss the application of genetic algorithms (GAs) in various other diversified fields.

Introduction to Genetic Algorithms in Search and Optimization

This chapter is all about introducing genetic algorithms in the search process, which are based on the theory of natural selection, genetics, and survival of fittest. By detailed understanding of the algorithm, one would be able to apply it in your respective field for optimization. At the end of this chapter, the reader will have acquired basic theory and working of these algorithms. Since genetic algorithms are used in diverse fields, the tone and language of this chapter is kept simple and casual for better understanding. Genetic algorithms in this chapter are applied through a hand calculation example. Genetic algorithms are basically mathematical calculations based on Darwin's theory of survival of fittest. This chapter gives a detailed understanding of the theory and working of genetic algorithms based on hand calculation examples. Comparison of genetic algorithms with other search procedures is also done.