C++ Neural Networks and Fuzzy Logic: Preface

Developing a Forecasting Model

There are many steps in building a forecasting model, as listed below.

1.  Decide on what your target is and develop a neural network (following these steps) for each

target.

2.  Determine the time frame that you wish to forecast.

3.  Gather information about the problem domain.

4.  Gather the needed data and get a feel for each inputs relationship to the target.

5.  Process the data to highlight features for the network to discern.

6.  Transform the data as appropriate.

7.  Scale and bias the data for the network, as needed.

8.  Reduce the dimensionality of the input data as much as possible.

C++ Neural Networks and Fuzzy Logic:Preface

Chapter 14 Application to Financial Forecasting

297

---

9.  Design a network architecture (topology, # layers, size of layers, parameters, learning paradigm).

10.  Go through the train/test/redesign loop for a network.

11.  Eliminate correlated inputs as much as possible, while in step 10.

12.  Deploy your network on new data and test it and refine it as necessary.

Once you develop a forecasting model, you then must integrate this into your trading system. A neural

network can be designed to predict direction, or magnitude, or maybe just turning points in a particular market

or something else. Avner Mandelman of Cereus Investments (Los Altos Hills, California) uses a long−range

trained neural network to tell him when the market is making a top or bottom (Barron’s, December 14, 1992).

Now let’s expand on the twelve aspects of model building: