How can I determine if I need a time series or a fitting neural network?

I have been working a little with Neural Networks but I sometimes get confused when needing to choose between a Fitting Neural Network (nftool) and a Time Series Neural Network (ntstool).

I know that Time Series NN has to do with time and sequences, but as its aim is to predict values, wouldn't it be similar to create a Fitting NN? I mean, creating a generalized Fitting NN that can be used with more inputs that the already given, so that a "predicted" result can be obtained as well.

Which is the difference between them?

Thank you very much in advance for your time and help!

Answer 1

[Greg Heath]

For a timeseries function points at neighboring times are highly correlated so that the autocorrelation function is nonrandom and/or inpnut and target points at neighboring times are highly correlated so that the corresponding crosscorrelation function is nonrandom.

For a fitting function only the input and target points at the same location are necessarily correlated. Random reordering of input/target pairs should make no difference,

Hope this helps.

Thank you for formally accepting my answer

Greg

[MRE]

Thank you very much for your answer. It was really helpful.