Algorithms used for forecasting
To properly make use of a machine learning technique such as forecasting, we need to understand how it works. Just making sure we have time-series data is not enough to get accurate predictions in Power BI. We need to understand the limitations and possibilities of the forecasting feature in Power BI to ensure that our results make sense. We'll first discuss how forecasting can be calculated, after which we will describe when we should use this feature in Power BI, based on the limitations and possibilities the underlying technique gives us.
The reason why some are skeptical of using an out-of-the-box feature such as forecasting in Power BI is because it is unclear what is actually happening under the hood. This kind of feature is often described as a black-box model – we can't see how the model is trained and how it creates predictions. Unfortunately, this is the intellectual property of Power BI and Microsoft and is, therefore,...
