Neural Networks or Artificial Neural Networks (ANNs) are computational models which simulate the connectivity of the neuronal structure of cerebral cortex and the brain learning patterns for certain computational tasks, such as machine learning, cognitive and pattern recognitions, etc.. Conventional computational models usually fare poorly in these areas.
Differing from Computational Neuroscience which offers in-depth study of the true complex biological neuronal functions in information processing in the brain, a neural network is more a simplified modeling technique or a set of algorithms in simulating the patterns of stimulations and repetitive learning of the brain by using interconnected parallel computational nodes as artificial neurons that are often organized into inputs, outputs and processing layers. Adaptive weights are used to simulate the connection strength between any two neuron nodes. Theses weights can be adjusted repeatedly by each “learning” cycle instead of being determined beforehand.
There are many college courses designated to the study of Neural Networks. In a simple sense, neural networks offer the possibility of continued learning and corrections in order to eventually fit the models closer to a particular function of brain by comparing the outcomes to certain reality. This is a huge deviation from the conventional computational models. Conventional models are deterministic with data and pre-defined instruction sets stored in memory for a centralized processor node to retrieve, compute and store in a sequential manner to generate outcomes. However the processing nodes for neural networks get information from input nodes or external signals to carry out simple weighted computations in parallel and the results are together presented as outcomes. The knowledge of a neural network is in the entire network itself instead of in any single node. Each computational cycle is almost a self-learning and reality-adjusting cycle like the way humans or animals generally learn.
Artificial Neural Networks in Trading
Neural networks have been a hot topic for traders for more than a dozen years. Indeed, this author has written many articles about neural networks and how they can be used in trading strategies.
The subject received a lot of attention because the technology had the ability to “learn” from past data and model problems where the underlying equations were unknown. Neural networks generalize well because they can give answers to new cases not used when the model was developed. They also handle noisy data well. These, and other, promises by everyone advocating neural networks in their heyday help the approach earn a large following.
Since then, however, we have learned that neural networks handle noise better than conventional statistical methods, but noise still needs to be a concern.
Since the 1990s, massive advances in inexpensive computing power have helped neural networks evolve. While the dot-com boom and bust dried up a considerable amount of speculative capital in the markets and interest in neural networks, today’s choppy markets are again inspiring traders to consider neural networks for an edge. In this three part series on neural networks, we will begin with an overview of the basics. Then, we will review research on using neural networks for trading. Finally, we will develop a case study using this technology in a real application.
