I have the following list of numbers (1 – 5 categories, first row) that I am trying to predict the next sequence for each column (total column sum = 6). There are over 1,500 lines of data in total. I have tried to make a column of the accumulated sum of each line and then use Linear Regression and Time Series Forecasting to predict the next accumulated sum and then subtract the previous number to predict the number (0,1,2,3,4) but I am not getting the results that I am after. How best can I forecast the next series of numbers using a non-Linear approach ?
You could use LSTMs with a sequence to sequence model perhaps (using e.g. Keras).
For example: x lines are input, the next line is output.
Just an idea, your question is not that clear.