Use of multiple models vs training a single model for multiple outputs

Question

So let's say I have data with numerical variables A, B and C.

I believe that the value of A has an effect on B. I also believe that A and B both have an effect on C. I don't think C has an effect on either A or B.

I want to use machine learning to predict A, B, and C. I obviously have A and B as training data, and I have other variables as training data too.

Do I simply create multiple models to predict all three, or is there a way to make one model predict them all if I just throw the entire dataset at it?

Answer 1

Do you have a data sample to answer better to your question?

For instance, are those variables related to time?

If yes, time series based models could be interesting like multi LSTM: LSTM Multi-class classification for large number of classes

If not, you could use a random forest regressor. https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html

My best advice is to start with a simple single model that makes predictions on A, B and C, and then try more complex ones.