LSTM time series prediction

hey every one I'm going to predict a big (1*50000) financial series. initially, I converted my data to (24*49976) with the purpose of 24 hours delays. now I've a matrix shape dataset can be used in ANFIS and neural network, easily. so the question is: is this true approach to build the input matrix? the second question is how can I use this (24*49976) dataset for deep learning algorithms exist in MATLAB; i means as i know Matlab Deep learning toolbox use for classification but my target is Continuous! so whats your ideas to solve my prediction problem? If you have an example or any guiding document please help me.

thanks in advance Abolfazl Nejatian

[Jaroslav Hook]

I need a 4 feature (most recent 10-20 lags ) predictor of a single value.

Which thenwork is the most suitable here ?

Answer 1

[Abolfazl Nejatian]

here is my code,

https://www.mathworks.com/matlabcentral/fileexchange/69506-time-series-prediction

this piece of code predicts time series data by use of deep learning and shallow learning algorithm.

best wishes

abolfazl nejatian