MATLAB dynamic neural network-time series prediction

I saw the time series prediction using dynamic neural networks on the matlat Chinese forum.

Http://www.ilovem http: // A http: // tlab.cn/thread-113431-1.html

(1) first basic knowledge needs to be known

Training data)

Validation Data)

Test Data)

However, I do not quite understand the three. Thank you for your explanation.

The following is an explanation of a Website:

Http://stackoverflow.com/questions/2976452/whats-the-diference-between-train-validation-and-test-set-in-neural-networks? Answertab = oldest # tab-top

The training and validation sets are used during training.

for each epoch for each training data instance propagate error through the network adjust the weights calculate the accuracy over training data for each validation data instance calculate the accuracy over the validation data if the threshold validation accuracy is met exit training else continue training

Once you're finished training, then you run against your testing set and verify that the accuracy is sufficient.

Training set: this data set is used to adjust the weights on the neural network.

Validation Set: this data set is used to minimize overfitting. You're not adjusting the weights of the network with this data set, you're
Just verifying that any increase in accuracy over the training data set actually yields an increase in accuracy over a data set that has not been shown to the network before, or at least the network hasn' t trained on it (I. e. validation data set ). if the accuracy
Over the training data set increases, but the accuracy over then validation data set stays the same or decreases, then you're overfitting your neural network and you should stop training.

Testing set: this data set is used only for testing the final solution in order to confirm the actual predictive power of the network.

(2) The second basic is to know

There are three types of time series prediction in the neural network toolbox that comes with Matlab.

A has an input x (t) and an output y (t ). For example: output relationship y (t) = y (t-1) + Y (t-2) +... Y (t-D) + x (t-1) + x (t-2) +... + x (t-d)

B is only output. For example, output y (t) = y (t-1) + Y (t-2) +... + Y (t-d)

C such as Y (t) = x (t-1) + x (t-2) + .. + x (t-d)

(3) Training Results

When using MATLAB Neural Networks for time series prediction, you must constantly adjust numbers of hidden neurons and number of delays D,

Finally, observe the two important graphs: plot error Autocorrelation and plot response. If the training is good,

Error autocorrelation is within the allowable range. If it is a type, there is also a plot inout-error correlation.

I have a qualitative judgment on these three types of videos, but I still don't understand the principle. Hope you can get some advice from others. Pai_^

(4) Prediction

The preliminary work has been done well, and our goal is to make predictions. Unfortunately, there is no Prediction Part in the video, but the video speaker is still

After realizing this problem, I posted a post to explain the prediction method to the audience.

Http://www.ilovematlab.cn/thread-132940-1-1.html

Http://www.ilovematlab.cn/forum.php? MoD = viewthread & tid = 216264.