Neural Networks and Stocks Trading

NB: We do not resell your papers. Upon ordering, we do an original paper exclusively for you.

NB: All your data is kept safe from the public.

Table of Contents

Introduction
How it works
Type of problem solved
Case study
Future of Neural Networks
Benefits of using neural networks
Drawbacks and Limitations
References

Introduction

An Artificial Neural Network (ANN) is a system that is based on how biological neural works, such as a brain, and the idea is transferred to information processing. That is, emulating how a neural system works. It is made up of many interconnected processing elements called neurons that are combined to solve a given problem. It tries to solve a problem from an example as people do. In each application, ANN has to be configured to be able to solve it.

How it works

ANN works with an example as human beings do. It does not work like a computer that follows a step-by-step procedure, but an example is identified which fits with the problem to be solved; otherwise, time will be wasted, or an incorrect solution will come out. The biggest disadvantage of an ANN is that how it works cannot be predicted because no procedure is followed. The input data comes from either the original data or out of another neuron. The signal is then taken to an activation function to produce an output. By using the activation function, an output of zero shows the input is less than zero, while an output of one shows the input is more than or is equal to zero. According to Ripley (1996), & subtracting the threshold from the weighted sum and comparing with zero is equivalent to comparing the weighted sum to the threshold&weights can be negative, which implies that the synapse has an inhibitory rather than excitatory effect on the neuron: inhibitory neurons are found in the brain.

The activation function is the most used method in ANN. The neurons are interconnected to get the input to transfer them to processing and then show the output. The input and output correspond to the eyes, hands of a human being, and the processing part is the brain which plays an internal role in the network. All three that is, input, hidden, and output, work together where the signal flows. The figure below shows a topology of how to input, hidden, and output is interconnected.

According to Haykin (1994), & the input variable values are placed in the input units, and then the hidden and output layer units are progressively executed. Each of them calculates its activation value by taking the weighted sum of the outputs of the units in the preceding layer and subtracting the threshold. The activation value is passed through the activation function to produce the output of the neuron. When the entire network has been executed, the outputs of the output layer act as the output of the entire network.

Type of problem solved

As noted by Carling (1992), the type of problem solved by ANN is not like for conventional computers that use algorithmic method, but it is unstructured. In this case, the problem is not understood, and there is no idea of how to solve it, and therefore no program can be developed to solve the problem. Hence, those problems are taken care of by Artificial Neural networks. To make ANN results more efficient, conventional computers are used for supervision.

Case study

Companies that use ANN include Insurance companies, banks, stock markets, among other institutions. The companies use ANN because they have some information and want to predict some unknown information. They use it for stock market prediction where the previous information is known and is used to determine future prices. ANN is also used also for credit assignment where information of the applicant is used to determine the risk on loan as bad or good.

Many institutions use it, and results do not differ from actual results unless there is no understanding of the problem. As stated by Ripley (1996), Many financial institutions use or have experimented with, neural networks for stock market prediction, so it is likely that any trends predictable by neural techniques are already discounted by the market, and (unfortunately), unless you have a sophisticated understanding of that problem domain, you are unlikely to have any success there either!

To avoid embarrassment, a neural network must be worked out in order to allow inputs and produce a real set of outputs. The best way is to use the knowledge of neural to know its strength by setting weights. Another method is to train it by inputting some values according to its pattern and allowing changes of its weight from its rules.

Future of Neural Networks

This is just the beginning of its application in financial institutions. It is predicted that more development is yet to take place due to its ability to use technology in the market. As stated by Patterson (1996), They are flexible, easy to integrate into a system, adapt to the data and can classify it in numerous fashions under extreme conditions. Therefore, to make it faster and efficient, hardware is being developed.

Benefits of using neural networks

Some of the benefits of using neural networks in stock trading and investments

Include:

neural networks require less training in formal statistics.
one is able to wholly identify complex nonlinear relationships between independent and dependent variables.
neural networks have the ability to detect all possible interactions between predictor variables,
It has the ability to produce multiples in training algorithms. (Gurney, 2003).

Drawbacks and Limitations

Neural Networks system can be very hard to use.
One has to fill the programs with settings where a minor error will translate to similar errors in the predictions.
It can be extremely hard to interpret results.
It becomes impossible to operate and analyze large neural networks due to the number of variables involved.
It is always impossible to comprehend a neural networks thinking even after generating a correct answer.
Further research is needed before neural systems acceptance in the industry. (Gurney, 2003).

Artificial Neural Networks is a more complicated way of solving problems because there are no procedures to follow, and hence not possible to know if the results are correct or wrong. When working with it, the user is required to be keen when inputting values to avoid getting wrong outputs. Also, it doesnt mean that the results obtained are 100 percent correct because of other factors that might not be provided for. Lastly, neural networks will be more efficient when integrated with conventional computing.

References

Carling, A. 1992, Introducing Neural Networks. Wilmslow, UK: Sigma Press.

Gurney, K. 2003, An Introduction to Neural Networks. London: CRC Press.

Haykin, S. 1994, Neural Networks: A Comprehensive Foundation. New York: Macmillan Publishing.

Patterson, D. 1996, Artificial Neural Networks. Singapore: Prentice Hall.

Ripley, B.D. 1996, Pattern Recognition and Neural Networks. London: Cambridge University Press.

Click Here To Order Now!