Forecasting extrapolation algorythm

[AlexeyRusinov]

Hello everyone,

Power BI Desktop offers an amazing forecasting feature, although I haven't found any descriptions of the algorithm that it uses. Which type of extrapolation it uses inside, where can I read about that? Also, are there any plans to make different types of forecasting available?

Any help would be much appreciated.  

[Anonymous]

Hi @AlexeyRusinov,

I haven’t found specific official article that describe what algorithm the forecasting feature uses in Power BI Desktop. However, there is detailed explanation about which algorithm Power View uses for forecasting in the following blog. 

https://powerbi.microsoft.com/en-us/blog/describing-the-forecasting-models-in-power-view/

As stated in the above blog, forecasting in Power View is based on an established suite of methods for time series prediction called exponential smoothing. Two versions of exponential smoothing are provided, one for seasonal data (ETS AAA), and one for non-seasonal data (ETS AAN). Power View uses the appropriate model automatically when you start a forecast for line chart, based on an analysis of the historical data.

This blog also applies to Power BI Desktop as Power BI Desktop started as these add-ins(Power Query, Power Pivot, Power View) for Excel.


Thanks,
Lydia Zhang

[jxjia]

It uses a modified version of Holt-Winters seasonal model.

For more detail, this article below provides a good level of detail about this methodology:

https://www.otexts.org/fpp/7/5

Cheers,

Gin

[tringuyenminh92]

Hi @jxjia,

I have sam concern with @AlexeyRusinov, We could not consult end-users use that feature when we dont know how it is implemented 😞 So could PBI team public some information to let us aware algorithms that forecasting feature using or be implemented?

[BhaveshPatel]

Please refer to this documentation at

https://powerbi.microsoft.com/en-us/blog/introducing-new-forecasting-capabilities-in-power-view-for-...