Starting December 3, join live sessions with database experts and the Microsoft product team to learn just how easy it is to get started
Learn moreShape the future of the Fabric Community! Your insights matter. That’s why we created a quick survey to learn about your experience finding answers to technical questions. Take survey.
12-20-2016 05:09 AM - last edited 12-20-2016 05:16 AM
Many faces of forecasting ...
Attached sample contains 4 fifferent ways to perform forecasting with exponential smoothing algorithm.
1) Using R-powered Custom Visual
2) Using R-visual
3) Using R-in-Power Query (data transform with R)
4) Using new "forecasting" option in good old Line Chart
Do not forget to install required R packages
The R code inside Power Query step is also attached (and also accessible in PBIX)
Why are the dates in the resultset skewed? Originally they are all first days of each month, as the first one in the screenshot below.
The algorithm only works with equally spaced dates. Therefore the original dates are resampled to become equally sampled in the range from first to last dates
Thanks very interesting, it demonstrates the potentialities of the product that many people can not understand at the moment