How to calculate Running totals for a constant value/month?

[Rck7]

Hi friends,

I would like your suggestion on calculating running totals for a given value/ month. 

Currently, I am calculating running totals for expected usage/month (i.e. 75000/month). 

I am trying to calculate running expected totals for my data in the data table(Oct-14 to Oct-17) where I came up with the below measure:

Expected Total = COUNTROWS (
SUMMARIZE (Records, Records[Date].[Year], Records[Date].[Month]))*75000 ,  which is giving me running total as 75000*37 months =2775000 (i.e. from27-Oct-14 to 26-Oct-17).

Instead I want my measure to calculate the expected running totals as 75000*36 months = 270000. 

How would I be able to modify my measure to calculate running toatals for 36 months? 

Kindly help!!!

Thank you. 

[v-jiascu-msft]

Hi @Rck7,

This could be a trick. The intervals from 27-Oct-14 to 26-Oct-17 are 36 months. But the records of month are 37 when you used COUNTROWS. Because October is counted one time more.

For example, from 2016-01-27 to 2017-01-26. There are 12 months intervals but 13 months of records. 2016-01 to 2016-12 and 2017-01. 

It seems you don't need concern the context. Maybe you can minus 1 directly. Or use function DATEDIFF instead.

DateDiff =
DATEDIFF ( DATE ( 2014, 10, 27 ), DATE ( 2017, 10, 26 ), MONTH )

Best Regards!

Dale

[Ashish_Mathur]

Hi,

Your qestion is not clear.  Share the link from where i can downlooad your file and also show your expected result.