Imagine you’re an information scientist at an enthralling little pet shop specializing in only five products: two varieties of cat food and three varieties of pet food. Your mission? To assist this small business flourish by accurately forecasting the weekly sales for every product. The goal is to supply a comprehensive sales forecast — total sales, in addition to detailed predictions for cat food and pet food sales, and even individual product sales.
The Data
You have got data on the sales of the different sorts of cat food A and B, in addition to the different sorts of pet food C, D, and E for 200 days. Lucky for us, the info is exceptionally clean, with no missing values, and no outliers. Also, there may be no trend. It looks like this:
Note: I generated the info myself.
Along with the person sales, we even have the aggregated sales for all cat food products, all pet food products, and all products. We call such a set of time series hierarchical time series. In our case, they respect the next sales hierarchy:
