Multivariate SKU Forecast Analysis
Multivariate forecasting predicts future demand for a SKU stream using demand and multiple attributes or factors that may influence demand. The use case does not rely solely on historical demand history, rather takes into account various external factors like weather data and economic factors, and harnesses the collective insights derived from these variables to generate future forecasts.
The information on this dashboard is grouped into the following widgets:
Widget | Description |
|---|---|
The SKU level error metrics generated from the Multivariate Machine Learning Forecast. | |
The SKU level error metrics generated from the Servigistics forecast. | |
The feature importance in descending order for the forecast generated with the value plot for a period range: • Average value of that feature across the period range • Red dot highlighting the global minima • Green dot highlighting the global maxima for each feature in their respective plot. | |
The graphical representation of demand and forecast for the selected SKU from the dropdown. |