How can I plot a confusion matrix to determine the prediction accuracy of a NAR predicted model?

I have used NAR (Nonlinear autoregressive model) to predict a time series.

Now I want to know the prediction accuracy of the neural network using a confusion matrix. How can I plot the same as I cant find any option to plot the confusion matrix in the given options. I can use plotconfusion(targets,outputs) but what would the targets and outputs be as you can refer from the figure where we have the actual time series and predicted time series.Kindly help

[RobF]

Are you sure the definition of a confusion matrix makes sense in this case? Usually you need classes to be able to define the confusion matrix. You might instead want to calculate the squared error as a measure for the correctness of the prediction.

[Kartik]

I have already calculated the mean square errors.I have sent the paper to a journal and the reviewer specifically asked for a confusion matrix for predicting the accuracy of neural network however i could not get the same thing you said. Can i get around this someway?

[Greg Heath]

Confusion matrices are appropriate for classifiers and pattern recognizers.

It appears that the reviewer is confused.

Hope this helps.

Greg