Is there a way to predict future year sales (5 years in the future) based on past performance and have it output in the form of a table instead of a bar graph or line graph? I can figure out how to find the 3-YR historical performance or 5-YR historical formance but how do I use such a formula to look forward 3 to 5 years into the future?
This is a pretty open ended question, because how are you predicting forward? You can easily jump around time frames using simple time intelligence questions but the reality is you need to have a good idea of how you’re going to predict some forward before you do that…
Here’s some ideas
So i have 5 years of registration (sales) data broken down by month. What I have done was create a measure that gives me the average of the last 3 completed years and one that gives me the average of the last 5 completed years. I would like to take the average performance of each month and use that specific average and the last completed academic year to predict the current year and perhaps scale that forward to predict up to 3-5 years in the future. I need to output the information in matrix format.
Is this possible to do?
Yes definitely.
I think this can easily to worked out bit by bit.
My recommendation here is to place this all into a table and work on each part of the logic one by one. To me it all sounds sequential so you can just calculate the individual parts and then bring them all together with some master formula that create the average or prediction going forward.
You can simply just use time intelligence functions to create the projection forward as per the examples above.
Some techniques covered below