Academic journal article Review of Business

Artificial Neural Networks Make Their Mark as a Powerful Tool for Investors

Academic journal article Review of Business

Artificial Neural Networks Make Their Mark as a Powerful Tool for Investors

Article excerpt


Artificial Neural Networks (ANNs) are fast becoming a powerful tool for investment forecasting and turning heads by achieving impressive results. For instance, an ANN-based system for investing in the U.S. Treasury Bond market was recently created and used to direct investments in that market for the years 1989-93. Over the five-year period, the ANN system generated a return on investment of 17 percent versus 14 percent for the prestigious Lehman Brothers Treasury Bond index over that same period. In another example, an ANN-based system was used to direct the investment of $10,000 in the S&P 500 index over a 25-month period. Results were spectacular, as the ANN was able to increase the fund to $76,034 over the 25 months.

The savvy investment manager simply cannot afford to ignore ANNs and the success they are achieving. This article explains what ANNs are, how they are being successfully used and how investors can learn more about this modern and very powerful tool to increase their own investment profits.

Forecasting in Investment Decisions

Forecasting future trends and events is a fundamental part of investing. Every investor who sinks hard-earned money into a particular investment (or alternatively decides to get out of an investment) is necessarily making a prediction about the future. He is making his best guess that a particular investment at a particular time bodes either more or less favorably for future returns than do other available choices. These decisions and the opposing conclusions reached by different individuals power the investment markets. For every buyer, there is a seller. Investors spend hundreds of dollars per year on newsletters attempting to get a jump on future market trends, and the odd guru who consistently out predicts the market is assured of an almost religious following.

In short, predicting the future directions of our financial markets is a large, important and very profitable business. Traditionally, investment forecasting has relied primarily on the expertise and "know-how" of experienced individuals. These individuals in turn have relied mainly on studying a wide variety of financial and economic data and applying their experience and intelligence to extrapolate future trends. While certain statistical and modeling tools, such as regression analysis, have helped, the process is still a very human one, often viewed as more art than science.

Enter the ANN

Over the past two decades, scientists have made significant strides in developing Artificial Neural Network systems that are now beginning to challenge traditional techniques of financial forecasting. These ANN systems can process an astonishing amount of data. They can search for and discover connective relationships and correlations among the various data in ways that humans cannot possibly match, simply by virtue of sheer microprocessing power. ANNs can use the data to make predictions, compare the actual results to the predicted results and then adjust their relationships or formulas so as to predict more accurately in the future.

In many cases, ANN systems are achieving market-forecasting results significantly better than those of major fund managers. These successes are happening in a wide variety of financial applications, from stock and bond market forecasting to bankruptcy failure prediction. Financial and investment institutions have taken notice, and neural network experts are being hired in increasing numbers by these concerns for the express purpose of developing and refining ANN-based market prediction systems. The day of computerized market forecasting is arriving, and the savvy investor will want to understand these systems and their strengths and weaknesses.

Introduction to ANNs

A neural network is basically an information-processing model designed to discover and track the relationships among various data sets autonomously. Since they were first developed in the 1950s, they have been called artificial intelligence systems, and indeed the analogy to human intelligence is a good one. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.