Tip |
In the case of prediction and especially when the series has few samples, it is not recommended to leave some samples for validation. There are several approaches that can be used to validate an ANN for prediction. One simple method to get a sense about the quality of the prediction is as follows:
|
Problem 1 |
Remove the last value (21.46) of the temperature series of the Simple Prediction section. Save the series with the name temperature63.csv. Perform the prediction analysis to compute the mse prediction analysis table and the value prediction analysis table. Using the results of the prediction analysis, predict the value of y[63]. If you do not want to perform these problems manually, you may write the PredValidation.lab file to let the computer do the validation process (you can use the function Matrix.Prediction to write this program.) |
Problem 2 |
Remove another value at the end of the temperature series (18.42). Save the series with the name temperature62.csv. Perform the prediction analysis to compute the mse prediction analysis table and the value prediction analysis table. Using the results of the prediction analysis, predict the value of y[62]. |
Problem 3 |
Keep removing values at the end of the series and complete the table below with the predicted value using for each case using an ANN. |
Series Length | Actual Value | Predicted Value |
63 | y[63]=21.46 | |
62 | y[62]=18.52 | |
61 | y[61]=18.17 | |
60 | y[60]=19.46 | |
59 | y[59]=20.27 | |
58 | y[58]=18.40 | |
57 | y[57]=15.14 | |
56 | y[56]=11.91 | |
55 | y[55]=11,91 | |
54 | y[54]=13.64 |
Problem 3 |
Using Microsoft Excel compute the mse between the actual and the predicted values using the 10 values in the table. In the first column of the sheet paste the Actual Value, in the second column paste the predicted value in the third column write the following formula =(A1-B1)*(A1-B1). The mse is the mean of the values in the third column. |