I am applying LSTM and GRU models to a financial problem where the dataset is a time series composed of 365 rows (365 continuous days) and 23 columns (my outcome is the closing price of a financial asset and the remaining 22 are other financial features).
I have two questions:
- Is it correct to use K-Fold Cross validation? Let's assume 5-fold, I am using non-continuous data to train and predict.
- If K-Fold is a good approach, should I divide each fold to be continuous or shuffled?