I am working on a new project I haven't much experience with and was looking for insight on where to begin and methods to use. I am trying to produce a demand forecasting model (or perhaps sales forecasting since I only have access to such data) to aid in inventory stocking levels for many products for one warehouse.
I have access to sales data on a few hundred products and only one store. I am constructing time series for the sales for each product and collecting them together in a dataset. I have done quite a bit of reading on the topic from both the web and here on cross-validated.
I am now in the spot where I should choose a method to begin with, and thought to begin with high demand products which are moving frequently (multiple sales on a weekly basis) and creating a forecasting model for solely for these products. I figure that creating an ARIMA model for each product will not be ideal and would like to obtain a forecast for each product with a single model. I have been mostly steering towards using a neural network model or employing a top-down hierarchical forecasting model.
Any insight is greatly appreciated. Thanks in advance.