When you set up a neural network time series regression, do you have to sort the cases in chronological order or is there a way to specify a time variable?
The data must be sorted in chronological order. When you specify the analysis it will ask you how many time steps need to be considered in making predictions, and this will take the data as presented, lag them according to that setting, and then use neural networks to predict the next step in the time series.
How do you decide how many steps to use in making predictions? Wouldn't you want to use all available data to give it as much information as possible for making the prediction?