Neural Network vs. Cycle Analysis to Predict the Stock Market

Many stock market experts consider that the last several years show the increasing of general market influence on the behavior of an individual stock. You could find the number around 40% for stock market risk-factor in old books. But now the general market influence is considered almost 85%. That is why it is very important to look at the stock market first and try to predict it before investing in any stock nowadays.

The stock market can be presented by S&P-500 index. It is possible to build its different statistical forecasts using historical data. The purpose of the research was to compare two statistical methods: one that based on Cycle Analysis, another - on Neural Network. We used price and volume data to train this particular Neural Network.

A picture below shows how actual 5-day performance (yellow line) differ from predicted performances by these two methods. The top half is the comparison of Neural Network prediction, bottom half - Cycle Analysis. Green bars mean buy signals, red - sell.



Three major conclusions:
1. Cycle Analysis prediction gives signals too early, Neural Network prediction - too late.
2. In average, Cycle Analysis prediction showed slightly better accuracy than Neural Network prediction for the last six months (from June 2009 to January 2010).
3. It it possible to get a superior prediction by combining these two methods.

---

The links where the models' descriptions can be found:
Cycle Analysis
Neural Network

---

© Alex Shmatov. Published with permission of the copyright owner. Further reproduction strictly prohibited without permission.