Forecasting with context
we have been able to use PAL Forecase smoothening algorithm to forecast value in a time series.
Ex: Feeding last 12 months usage to forecast next 6 months.
Now, we want to extend and contexualize the forecasting with parameters like region, usage of specific resource type etc.
- Demand Forecast of resources (equipment types/personnel types) for future jobs
- Use past months historical jobs to predict demand for each resource types
- Like which customer will have more demands for which Resource Type
- Revenue and margin forecast based on Field tickets & margin analysis for past months
Can someone provide any pointers on how to extend PAL forecasting algorithms to be used with additional parameters ?
Antoine CHABERT replied
You might also want to have a look to the time series capabilities of Automated Analytics. You have the possibilities to use extra-predictors and through scripting you can create many models, segmented according to region, product...
Please see the two tutorials from the one & only Former Member here:
Thanks & regards