I have a task of forecasting the number of calls with a 5-minute interval for the next 14 days. We have data for the last 4 years with a 5-minute interval, and there are only gaps for one month. I plan to use a neural network model to solve this task. However, I'm not sure about the best approach:
Should I first forecast the number of calls for the next day and then use these forecasts to predict the number of calls in intervals for the next 14 days? Or is it more advisable to directly predict the number of calls in intervals for the next 14 days?