C++ Neural Networks and Fuzzy Logic: Preface

1.  Take the predicted scaled target value and calculate, the result value as Result = (Scaled

target/0.02) − 14.8

2.  Take the result from step 1 (which is the percentage change 10 weeks from now) and calculate

the projected value, Projected S&P 10 weeks from now = (This week’s S&P value)(1+ Result/100)

Only the Beginning

This is only a very brief illustration (not meant for trading !) of what you can do with neural networks in

financial forecasting. You need to further analyze the data, provide more predictive indicators, and

optimize/redesign your neural network architecture to get better generalization and lower error. You need to

present many, many more test cases representing different market conditions to have a robust predictor that

can be traded with. A graph of the expected and predicted output for the test set and the training set is shown

in Figure 14.6. Here, the normalized values are used for the output. Note that the error is about 13.9% on

average over the test set and 6.9% over the training set. You can see that the test set did well in the beginning,

but showed great divergence in the last few weeks.

Figure 14.6

  Comparison of predicted versus actual for the training and test data sets.

The preprocessing steps shown in this chapter should serve as one example of the kinds of steps you can use.

There are a vast variety of analysis and statistical methods that can be used in preprocessing. For applying

fuzzy data, you can use a program like the fuzzifier program that was developed in Chapter 3 to preprocess

some of the data.

What’s Next?

There are many other experiments you can do from here on. The example chosen was in the field of financial

forecasting. But you could certainly try the simulator on other problems like sales forecasting or perhaps even

weather forecasting. The key to all of the applications though, is how you present and enhance data, and

working through parameter selection by trial and error. Before concluding this chapter, we will cover more

topics in preprocessing and present some case studies in financial forecasting. You should consider the

suggestions made in the 12−step approach to forecasting model design and research some of the resources

listed at the end of this chapter if you have more interest in this area.

C++ Neural Networks and Fuzzy Logic:Preface

Only the Beginning

315

---

Previous Table of Contents Next

Copyright ©

 IDG Books Worldwide, Inc.

C++ Neural Networks and Fuzzy Logic:Preface

Only the Beginning

316

---