PredictorXL: Example Stock Price Prediction
You have a table with historical stock price data, two technical analysis functions and you would like to get a prediction for tomorrow's Closing price:
Step 1: You click on PredictorXL from the Excel menu:
After the launch of the program you will see the PredictorXL form.
Step 2: Using the mouse button choose the range of numerical data (Inputs) of the table that you want to use and which is influencing on the outputs and choose the range for the Outputs.
Step 3: Set the following parameters: 1) Learning rate, which influences the speed and quality of training; 2)
Lag - number of predicted values. (If set to 1, then it will predict only tomorrow's value, if set to 2 - it will predict for tomorrow and the day after tomorrow.);
3) Error - error value, when the neural net will automatically stop training;
4) Momentum, which influences the speed and quality of training.
Step 4: Click the New button, Start Training button and Show Process button. The PredictorXL dialog will transform and will show a training graph.
Blue line shows real values, red line shows predicted values:
Step 5: When the training is complete, you will see a notification box. Click the Predict button and the following dialog will appear:
This produces the following prediction result:
You received the result and you want to know what PredictorXL would predict if you launched it yesterday.
So, you need to decrease the Inputs and Outputs array:
Step 6. Click the New button and Start Training button:
Step 7: When the training is complete, you will see a notification box. Click the Predict button and the following dialog will appear:
In the current example the resulting prediction error is only $0.69.