Choose end date for ML forecast? Export ML forecast (insight) values to table visual?

Hi there,

I have two questions relating to the ML insights forecasting capabilities in quicksight.

  1. Is it possible to manually select the end date for an ML forecast?

  2. Is it possible to see forecasted values in a table visual in quicksight?

In my dataset I am interested in the cumulative count of transactions received each month. In the current month I would like to see the forecasted, cumulative total for the end of the month, based on the trend of transactions so far this month.
I would also like to see this forecasted, end-of-month total, for several different organisational groups simultaneously. If it is possible, I would like to see these forecasted values in a pivot table, and thereby be able to compare across the different groups in that table.


RE: Q1):
In some cases, the insights tab on the left hand ribbon in the quicksight analysis interface has provided the end of month forecast in an insight widget:

In the above image I can see the forecasted total for the end of July. This is great!

Unfortunately, in other cases, the date for the forecast is not the last day of the month:

In the above image, the date of the forecast is for the 7th of November, when I would like to see the forecasted value for the 31st of October.

If anyone knows whether it is possible to:

a) specify the end date for the ML forecast.


b) export these forecasted values into some field that could be included in a table visual

I would be very interested in hearing your response!

Many thanks!

a) Have you tried adding a filter to your visual before you make the forecasted insight? Trying to exclude any additional day? I don’t know if this will work but curious to see what changes if anything.

b) It doesn’t look like you can export it. I will mark this as a feature-request as it makes sense.

Hi @Max,

Thank you for your response!

RE a) I like the idea of adding a filter on the insight. Unfortunately, I just tested it, and it didn’t work for me.

RE b) Thank you!