Statistical forecasting methods — does Fabric Plan support ARIMA or ETS?

Started by Marcus Delacroix · 1 year, 1 month ago · 3 replies · 1 view

Marcus Delacroix Participant

1 year, 1 month ago

We want to use time-series statistical methods to generate baseline forecasts which planners then adjust. Does Fabric Plan have a statistical forecasting engine, or should we generate forecasts externally and import them?

3 Replies

Sort

Raj Subramaniam

Microsoft Fabric Consultant · CloudBridge Solutions · 1 year, 3 months ago

Fabric Plan's built-in forecasting is driver-based and method-based — no ARIMA/ETS engines natively. For statistical forecasts: generate in Azure ML or Fabric Notebooks, write to a Fabric SQL table, then surface as a 'Statistical Forecast' measure in your semantic model.

Up

0

David Thornton

Director of Enterprise Analytics · Stellar Manufacturing · 1 year, 3 months ago

The architecture of running the statistical engine in Fabric and exposing the output to Plan via the semantic model is elegant. The planner doesn't need to know about the model — they just see a starting point and adjust.

Up

0

Omar Hassan

Data Engineering Lead · InfraCore MENA · 1 year, 3 months ago

We do exactly this — Python notebooks in Fabric generate Prophet forecasts for our demand planning, write to SQL, and Fabric Plan's Planning sheet shows the statistical baseline alongside the planner's override. Best of both worlds.

Up

0